A method and device for intelligent selection of factory automation parts

Abstract

The present invention relates to the technical field related to factory manufacturing, and specifically includes a method and device for intelligent selection of factory automation parts, including: establishing an intelligent selection platform, including input, adjustment, and fuzzy recognition ports; the platform integrates user input data to generate application scenarios, and corrects and compensates within a preset time; activates fuzzy recognition, collects fuzzy and historical data, parses and reconstructs a database, matches parts, and completes intelligent selection, which solves the technical problems of a wide variety of parts involved in factory automation and a lack of an adaptive optimization mechanism, resulting in poor flexibility in part selection and affecting the overall processing and manufacturing efficiency of the factory, and integrates data from different sources, realizes efficient automatic selection through a fuzzy recognition port and a historical matching call function, and performs automatic compensation and correction according to a priority database to improve the accuracy of selection.

Description

Technical Field

[0001] The present invention relates to the technical field related to factory manufacturing, and in particular to a method and device for intelligent selection of factory automation parts. Background Art

[0002] There are many types of parts involved in factory automation, covering multiple fields such as machinery, electronics, sensors, etc. With the continuous updating of technology, the performance and functional requirements of parts in the manufacturing industry are becoming more and more diversified, and the selection process has become complicated and time-consuming. At present, many factories rely on system knowledge when selecting parts, but because different users have different requirements for production requirements, product design and project background, there is a lack of effective real-time optimization mechanism when facing fuzzy data and user adjustment feedback, which affects the overall processing and manufacturing efficiency of the factory.

[0003] In summary, the existing technology has technical problems such as a wide variety of parts involved in factory automation and a lack of adaptive optimization mechanism, which leads to poor flexibility in parts selection and affects the overall processing and manufacturing efficiency of the factory. Summary of the invention

[0004] The present application provides an intelligent selection device for factory automation parts, aiming to solve the technical problems in the prior art that there are many types of parts involved in factory automation, lack of adaptive optimization mechanism, resulting in poor flexibility in part selection and affecting the overall processing and manufacturing efficiency of the factory.

[0005] In view of the above problems, the technical solution to implement this application is:

[0006] In one aspect, the present application provides a method for intelligent selection of factory automation parts, wherein the method comprises: establishing an intelligent selection platform, wherein the intelligent selection platform is configured with an input port, an adjustment port, and a fuzzy recognition port;

[0007] The input port is used to receive the user's input data, which includes design drawings, project specifications, and production requirements. The input data is integrated and identified through the scenario establishment channel built into the intelligent selection platform, and the application scenario is output through the adjustment port. The application scenario is the result of the integrated identification;

[0008] Reading user adjustment data of the adjustment port within a preset time window, and correcting and compensating the application scenario with the user adjustment data;

[0009] Activate the fuzzy recognition port to collect the user's fuzzy data, where the fuzzy data includes fuzzy input data and historical matching call data;

[0010] After performing fuzzy analysis on the fuzzy input data in the fuzzy data, the matching database is reconstructed with the historical matching call data. The fuzzy analysis results and the application scenarios after correction and compensation are used as matching data. The component matching of the reconstructed matching database is performed, and the intelligent selection is completed according to the component matching results.

[0011] On the other hand, the present application provides a device for intelligent selection of factory automation parts, wherein the device comprises: a port configuration module for establishing an intelligent selection platform, wherein the intelligent selection platform is configured with an input port, an adjustment port, and a fuzzy recognition port;

[0012] The data receiving module is used to receive the user's input data through the input port, and the input data includes design drawings, project specifications, and production requirements. The input data is integrated and identified through the scenario establishment channel built into the intelligent selection platform, and the application scenario is output through the adjustment port. The application scenario is the result of the integrated identification;

[0013] A correction and compensation module, used to read user adjustment data of the adjustment port within a preset time window, and to correct and compensate the application scenario with the user adjustment data;

[0014] A data collection module, used to activate the fuzzy recognition port and collect the user's fuzzy data, wherein the fuzzy data includes fuzzy input data and historical matching call data;

[0015] The intelligent selection module is used to perform fuzzy analysis on the fuzzy input data in the fuzzy data, reconstruct the matching database with historical matching call data, use the fuzzy analysis results and the application scenarios after correction and compensation as matching data, perform component matching of the reconstructed matching database, and complete intelligent selection according to the component matching results.

[0016] In summary, one or more technical solutions provided in this application solve the technical problems that the wide variety of components involved in factory automation and the lack of adaptive optimization mechanism lead to poor flexibility in component selection and affect the overall processing and manufacturing efficiency of the factory. It integrates data from different sources, realizes efficient automatic selection through fuzzy recognition ports and historical matching call functions, and performs automatic compensation and correction according to the priority database, thereby improving the technical effect of the accuracy of selection. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 A flow chart of a method for intelligent selection of factory automation parts is provided for this application;

[0018] Figure 2 A structural schematic diagram of a factory automation component intelligent selection device is provided for this application.

[0019] Description of reference numerals: port configuration module M100, data receiving module M200, correction and compensation module M300, data acquisition module M400, intelligent selection module M500. DETAILED DESCRIPTION

[0020] Embodiment 1

[0021] The present application is described in detail below with reference to the accompanying drawings. Figure 1 As shown, the present application provides a method for intelligent selection of factory automation parts, wherein the method comprises:

[0022] S1: Establish an intelligent selection platform, which is equipped with an input port, an adjustment port, and a fuzzy recognition port; S2: Use the input port to receive user input data, which includes design drawings, project specifications, and production requirements. A channel is established through a scenario built into the intelligent selection platform to integrate and identify the input data, and the application scenario is output through the adjustment port. The application scenario is the result of the integrated recognition.

[0023] Specifically, the input port is the input interface of the intelligent selection platform, through which users upload various forms of input data, such as drawings, project specifications or production requirements; the adjustment port is facing the user end, and users can modify or optimize the application scenario through the adjustment port; the fuzzy recognition port is used to receive inaccurate or fuzzy data input; the scenario establishment channel is used to integrate the user input data into specific application scenarios to help the selection process.

[0024] Execution steps: Users submit design drawings, project specifications and production requirements through the input port of the intelligent selection platform. Generally, the input design drawings include the appearance, structure and geometric dimensions of the equipment, the project specifications include functional requirements, and the production requirements involve equipment production efficiency, carrying capacity, etc.; input data into the scene to establish a channel, the intelligent selection platform integrates all input data, extracts geometric features from the design drawings, performs natural language processing on key information in the specifications, digitizes the production requirements, and then generates application scenarios. The application scenarios include specific details such as the material requirements and usage environment of the robotic arm.

[0025] After data identification and integration are completed, the intelligent selection platform outputs an application scenario through the adjustment port. Users can view the preliminary selection suggestions generated by the intelligent selection platform, such as the specifications of each component. The fuzzy identification port in the intelligent selection platform can process the fuzzy data input by the user, and analyze the fuzzy input and the historical matching data of the platform so that users can obtain more flexible selection suggestions. For example, the user inputs a fuzzy requirement, such as "faster speed". Through fuzzy logic combined with the user's historical usage data, an adaptive high-speed conveyor belt and a faster response application scenario are selected. At the same time, the application scenario is the result of integrated identification, and the application scenario is output through the adjustment port. By optimizing the selection process through the above steps, precise adjustments can be made according to specific requirements, and flexible matching can be provided through fuzzy input.

[0026] S3: within a preset time window, read the user adjustment data of the adjustment port, and use the user adjustment data to correct and compensate the application scenario; S4: activate the fuzzy recognition port to execute the user's fuzzy data collection, and the fuzzy data includes fuzzy input data and historical matching call data; S5: after performing input fuzzy analysis according to the fuzzy input data in the fuzzy data, reconstruct the matching database with the historical matching call data, use the fuzzy analysis results and the corrected and compensated application scenario as matching data, execute component matching of the reconstructed matching database, and complete intelligent selection according to the component matching results.

[0027] Specifically, the preset time window refers to the time period within a specific time period during which the user is waiting to input adjustment data; the adjustment data is the modification or supplementary information made to the scenario by the user after outputting the application scenario, and the adjustment data is used to further optimize the selection process; correction compensation is the modification or compensation of the original scenario based on the user's adjustment data to ensure that the output result better meets user needs; fuzzy data is incomplete or unclear user input data, including fuzzy input data and historical matching call data, which is used to process imprecise requirements, among which the historical matching call data is expressed as a call constraint interval to characterize the user's intention to use historical data.

[0028] Execution steps: After the intelligent selection platform generates a preliminary application scenario, it enters a preset time window period and waits for users to adjust the application scenario. Users input their modification opinions or additional requirements through the adjustment port. The adjustment data can be modifications to component specifications, material changes, or adjustments to other details. Commonly, users find that the motor power in the preliminary selection recommendation is insufficient, and change the motor power to a higher specification through the adjustment port; read the user's adjustment data, and apply the adjustment data to the preliminary application scenario for correction and compensation. For example, if the user adjusts the motor power, it is also necessary to adjust the power consumption, heat dissipation requirements and other related parameters in the scenario based on this information to compensate for changes in the application scenario.

[0029] Activate the fuzzy recognition port and start receiving fuzzy data input by the user. The fuzzy data input by the user may include unclear requirements, such as "higher precision". The historical matching call data will be retrieved and combined with the previous user's selection preferences or industry standards to further optimize the matching process. Specifically, similar requirements are identified from historical records, and a series of precision parts are recommended to users in combination with historical data on accuracy levels. The fuzzy input data is fuzzily parsed and the fuzzy input data is gradually converted into specific parameters or constraints using fuzzy logic algorithms. For example, "higher precision" will be parsed into a specific tolerance range or accuracy level. The intelligent selection platform combines historical call data to find past parts matching records that meet user requirements. For example, the material selection of the robotic arm and the specifications of each component will parse "higher precision" into a tolerance range of ±0.01mm, and locate and search for relevant historical records.

[0030] Through the fuzzy analysis results and historical matching call data, the parts matching database is reconstructed to ensure that the matching solutions in the database are not only based on the current input data, but also take into account user preferences and needs, making the data in the database more targeted and able to provide more personalized parts matching suggestions. For example, "higher precision" is parsed as a tolerance range of ±0.01mm. After reconstruction, the database adds parts with higher precision for user needs and eliminates low-precision parts that are not suitable for the application scenario; then, parts matching is performed based on the reconstructed matching database, comprehensively considering the fuzzy analysis results and the application scenarios after correction and compensation, and performing matching calculations for parts selection, outputting parts matching solutions that meet user needs, and completing intelligent parts selection. Through the above steps, one-stop parts selection is carried out to improve the overall efficiency of the factory.

[0031] Furthermore, the integration and identification of input data through the scene establishment channel built into the intelligent selection platform, the method of the present application also includes:

[0032] The input data is analyzed through the preprocessing layer to establish the data analysis results; the data type analysis of the input data is performed using the data analysis results, and the first focus factor of the recognition layer is established based on the data type analysis results; the data richness analysis of each data type is performed using the data analysis results, and the second focus factor of the recognition layer is established based on the data richness analysis results; the recognition accuracy of the recognition layer is configured according to the first focus factor and the second focus factor, and the configured recognition layer is used to identify the data analysis results and establish an application scenario.

[0033] Specifically, the preprocessing layer is mainly used to parse and organize input data, and to preliminarily classify and clean different types of data; data type analysis refers to classifying the parsed data and identifying its data type, such as text, numerical or image data; the recognition layer is responsible for configuring the recognition algorithm according to the type and richness of the input data, which is used to further analyze the data and establish application scenarios; the first focus factor refers to the key data features or factors that the recognition layer focuses on based on the results of the data type analysis, which is used to guide the intelligent selection platform to adjust the recognition accuracy; the second focus factor refers to determining the data scope or level that the intelligent selection platform focuses on based on the data richness analysis; recognition accuracy refers to the accuracy and detail when parsing data, and is adjusted according to the data characteristics.

[0034] Execution steps: Parse the user input data through the preprocessing layer. The input data is in various forms, such as design drawings (image data), project instructions (text data or graphic data), production requirements (text data), etc. The input data is classified and extracted through the preprocessing layer for subsequent analysis; perform data type analysis on the parsed data. Different types of data (such as images, text, and numbers) have different characteristics. Set corresponding attention factors according to the data type to ensure that the data processing method adapts to its characteristics. Specifically, the image data in the design drawings obtains graphic parameters through geometric shape feature analysis, and the instruction manual text extracts key design requirements through natural language processing (NLP). The first attention factor will include geometric features, material type, operating temperature, etc.

[0035] According to the amount and complexity of the parsed data, the data richness analysis is performed. The higher the data richness, the more attention factors are needed to ensure that the recognition process can fully consider various features. Specifically, if the input design drawing contains detailed geometric data of multiple sub-components, and the manual describes the production requirements under various working conditions, the second attention factor is established according to the complexity and diversity of the data, such as the number of components, the range of environmental changes, etc.; Based on the analysis results of the first attention factor and the second attention factor, the corresponding recognition accuracy is configured for the recognition layer. The higher the recognition accuracy, the more detailed the analysis of the data, and the more accurately the key points in the user's needs are identified; When processing complex design drawings, the accuracy of geometric feature recognition is improved to ensure that the details of the components can be accurately captured, and for the text in the manual, the focus is on extracting specific process requirements; The configured recognition layer is used to further identify the data parsing results, and various types of data are integrated to form application scenarios. For example, the application scenarios suitable for user needs are determined by combining the geometric information in the design drawings, the production requirements in the project manual, and the specifications of the parts that match the user's needs. Through the above steps, the recognition accuracy is optimized to determine the application scenario by comparing the first attention factor and the second attention factor, and data of different types and richness are efficiently processed.

[0036] Furthermore, the method of the present application further comprises:

[0037] The data analysis results corresponding to the design drawings are synchronized to the image recognition layer, and the geometric shape features and material property features are used to extract data features to establish the drawing analysis results; the data analysis results corresponding to the project specifications are synchronized to the natural language processing layer, and the natural language processing layer is used to extract key text information to establish the text analysis results; the data analysis results corresponding to the production requirements are synchronized to the numerical processing layer, and the numerical information is digitized using the numerical processing layer to establish the digital analysis results, wherein the image recognition layer, natural language processing layer, and numerical processing layer are all sub-processing units of the recognition layer, and the recognition accuracy is configured based on the first focus factor and the second focus factor; feature fusion is performed according to the drawing analysis results, text analysis results, and digital analysis results to complete the recognition.

[0038] Specifically, the image recognition layer is used to process modules of image data such as design drawings, and extracts useful information from images by identifying features such as geometric shapes and materials; the natural language processing layer is used to extract key information from project specifications or other text data; the numerical processing layer is used to process modules of numerical data, and is responsible for digitizing the numerical information in production requirements and extracting valid numerical information.

[0039] Execution steps: Pass the data of the design drawing to the image recognition layer for processing. The image recognition layer will use geometric shape features (such as the component shape in the drawing) and material property features (such as the material type marked in the drawing) to extract the key information of the drawing. The design drawing contains the geometric design and material description of a device. The image recognition layer identifies the component structure of the device by extracting the geometric shape information, and determines that the device uses steel and aluminum alloy based on the material properties.

[0040] The text information in the project specification is passed to the natural language processing layer, which uses natural language processing technology to extract key design requirements, production requirements or constraints in the specification. The project specification describes the production capacity requirements and process flow of the equipment. The NLP layer extracts key information, such as "the processing speed must reach 100 pieces per minute" or "the ambient temperature is required to be between -10°C and 40°C"; the numerical data related to the production requirements is synchronized to the numerical processing layer, which performs digital processing and converts the numerical information into a recognizable format. The production requirements include specific information such as production line speed, temperature control, pressure setting, etc., to facilitate subsequent matching.

[0041] According to the data type and richness analysis, the recognition accuracy of the image recognition layer, NLP layer and numerical processing layer are configured respectively. The first focus factor and the second focus factor ensure that different data are analyzed with different accuracy to adapt to the complexity and importance of the data. Specifically, for complex design drawings, the recognition accuracy of the image recognition layer is increased to ensure that every detail is accurately captured; for important text paragraphs in the manual, the NLP layer will increase the parsing depth according to their importance.

[0042] The analysis results of the image recognition layer, natural language processing layer and numerical processing layer are feature fused, and the feature information from different data sources is integrated to form an application scenario. At this time, the user needs are understood more comprehensively and the selection of parts is recommended; the image recognition layer extracts the geometric structure, the NLP layer extracts the production requirements, and the numerical processing layer extracts the key numerical parameters; after the key numerical parameters are fused, the specific performance parameters that the equipment needs to have are obtained, and suitable parts are recommended based on the specific performance parameters, such as the recommended motor type, the required number of sensors, etc. In the above steps, multi-source data analysis (images, texts, and numerical values) is used to ensure that key information is extracted from different data sources, and then intelligent component selection in complex scenarios is achieved through feature fusion.

[0043] Furthermore, the method of reconstructing the matching database with historical matching call data also includes:

[0044] Read the user's account data and establish the user's usage database based on the account data; parse the historical matching call data to obtain the user's single call interval, which is the user's historical data usage preference interval; establish a general database, fuse the general database and the usage database based on the single call interval, and use the data fusion result as the reconstructed matching database.

[0045] Specifically, account data refers to personal information, operation records, preferences and other data generated by users in the intelligent selection platform, which are used for personalized recommendations and matching; the usage database is a database established based on user account data, which saves user selection preferences, historical operations and other information; historical matching call data is a data record of historical queries, matches and usage of users in the intelligent selection platform, reflecting user habits and needs; a single call range refers to the data usage range of a user in a certain operation or matching process, representing the specific demand range of the user in a selection; the general database is basic data such as standard components, general matching algorithms, etc. stored in the intelligent selection platform, which is used for data matching of all users.

[0046] Execution steps: Read the user's account data, including the user's personal information, historical operation records and preferences. According to the account data, establish the user's usage database. The user's usage database mainly records the user's personalized needs and habits, providing a basis for subsequent matching; by analyzing the user's historical matching call data, extract the specific data usage range of each call made by the user, that is, the single call range. The single call range usually reflects the user's demand preferences in a specific selection task.

[0047] The intelligent selection platform also has a general database that stores standard parts data, recommendation models and other information. The data in the general database is integrated with the data in the user database. The key to integration lies in the single call interval, that is, the specific needs and preferences of users in historical calls are given priority, and the user's historical data usage preference interval is determined. The single call interval is the user's historical data usage preference interval; the general database and the usage database are integrated based on the single call interval. The fused data fusion result will be used as the reconstructed matching database, which contains the standard information of the general database and combines the user's personal needs and historical preferences, so as to achieve more accurate and personalized parts recommendations. For example, when matching parts, products that are consistent with the user's past choices and preferences are recommended first, helping users find suitable parts more quickly. In the above steps, the matching database is reconstructed through historical matching call data and user account information, and the dynamically reconstructed matching database is constructed to more accurately reflect the user's actual needs, thereby improving the matching accuracy and efficiency of the intelligent selection process.

[0048] Furthermore, the correction and compensation of the application scenario by using the user adjustment data, the method of the present application also includes:

[0049] The adjustment factors of the user adjustment data are identified, and the adjustment factor identification results are established; the adjustment factor identification results are used for priority database matching, and the priority database includes a first priority factor and a second priority factor, the first priority factor is an immediate compensation factor, and the second priority factor is a delayed compensation factor; and application scenario correction compensation is completed according to the priority database matching results.

[0050] Specifically, user adjustment data refers to the personalized adjustments made by users to input results or matching scenarios when using the intelligent selection platform, which generally come from changes in user needs or user dissatisfaction with the initial matching results; the priority database refers to the classification of adjustment factors into immediate compensation factors and delayed compensation factors based on the urgency and impact time of the adjustment. Among them, immediate compensation factors are factors that need to be adjusted immediately, such as immediate changes in user needs in the current scenario; delayed compensation factors are adjustment factors that can be processed after a certain period of time, such as long-term changes in demand or preferences.

[0051] Execution steps: When the user adjusts the matching results, these adjustments will be analyzed to identify the factors behind the adjustments. For example, when the user changes the material or specification of a component, it is necessary to identify the reason why the user made the adjustment, which is generally due to changes in production requirements or adjustments in user preferences. After identifying the adjustment factors, the adjustment factors will be matched with the priority database, which contains two types of factors: immediate compensation factors and delayed compensation factors. Among them, immediate compensation factors may be problems that must be dealt with immediately in the current task, while delayed compensation factors are factors that can be optimized and adjusted in subsequent operations. According to the priority matching results, the current application scenario is corrected and compensated. If it is an immediate compensation factor, the matching result will be readjusted immediately to ensure that the current user's needs are met; if it is a delayed compensation factor, the delayed compensation factor will be recorded and gradually optimized and adjusted in subsequent matching scenarios. Through the above steps, the application scenario is corrected and compensated according to the user's adjustment data, the user's adjustment factors are identified, and the immediate compensation (immediate compensation ensures the smooth progress of the current task) and delayed compensation (gradually optimized over a long period of time) are processed through the priority database to adapt to the user's changing needs.

[0052] Furthermore, the present application method also includes:

[0053] Establish a balance adjustment weight constraint as follows:

[0054] in, Adjust weights for users, Adjust the weights for the system, Adjust frequency for users, Characterizes the user-adjusted sensitivity coefficient, is the balance point, representing the standard user adjustment frequency, Normalized value for user-adjusted amplitude, is the influence coefficient of the adjustment amplitude; and the user adjustment data is constrained by balancing the adjustment weight constraint to correct and compensate for the application scenario.

[0055] Specifically, the user adjusts the weights Refers to the weight given to the user's adjustment behavior when the user adjusts the matching results. The larger the weight, the more priority the user's adjustment needs will be given. The system adjusts the weight Refers to the evaluation weight of the intelligent selection platform itself on the current matching result. If the intelligent selection platform has high confidence in the matching result, the system adjustment weight will be larger, thereby reducing the impact of user adjustment; User adjustment frequency Indicates the frequency of adjustments made by users when using the intelligent selection platform. If users make adjustments frequently, the sensitivity to user adjustments will increase. Sensitivity coefficient Used to measure the sensitivity of user adjustments. Users who make frequent adjustments will have a higher sensitivity coefficient. Balance point Indicates the adjustment frequency of the standard user, which is a reference value determined by a large amount of historical data. If the user's adjustment frequency exceeds or falls below the standard user's adjustment frequency, the matching strategy will be adjusted accordingly; the normalized value of the user's adjustment amplitude Refers to the normalization of the user's adjustment range to facilitate comparison within different adjustment ranges; the influence coefficient of the adjustment range It is used to measure the impact of the user's adjustment range on the matching results. If the adjustment range is large, the impact coefficient will increase.

[0056] Execution steps: Usually, the sum of user-adjusted weight and system-adjusted weight is set to 1; user-adjusted weight During multiple adjustments, the intelligent selection platform records the user's adjustment behavior, such as how many times the user made adjustments and how much each adjustment was made. The intelligent selection platform classifies the multiple adjustments and determines which adjustments are effective immediately (immediate compensation factors) and which adjustments require long-term observation to take effect (delayed compensation factors). Then, the user's adjustment weight is calculated based on factors such as the user's adjustment frequency and adjustment range, and substituted into the weight. The system adjustment weight is obtained from the , and the application scenario is adjusted accordingly. Through the above mechanism, the intelligent selection platform can quickly respond to the actual needs of users, while ensuring overall stability and reliability.

[0057] Furthermore, the present application method also includes

[0058] Acquire user selection feedback, classify the selection feedback, and establish a feedback classification data set, wherein the feedback classification data set includes a positive feedback data set and an abnormal feedback data set; use the feedback classification data set to establish timing compensation, and perform platform self-optimization of the intelligent selection platform through the timing compensation.

[0059] Specifically, selection feedback refers to the feedback information given by users after using the intelligent selection platform to select parts, including satisfaction with the selection results, usage experience, etc.; feedback classification refers to the classification based on the nature of user feedback, which is generally divided into two categories: positive feedback and abnormal feedback. Among them, the positive feedback data set contains all users' positive evaluation data on the part matching results of the intelligent selection platform, indicating that the part matching results of the intelligent selection platform meet user needs; the abnormal feedback data set contains users' negative evaluation data on the part matching results of the intelligent selection platform, indicating that the recommended results do not meet user needs and need to be adjusted; timing compensation refers to a time-based feedback analysis mechanism, which is used to dynamically adjust platform parameters according to feedback from different periods, so that the intelligent selection platform can make more reasonable recommendations in subsequent operations.

[0060] Execution steps: After the user completes the component selection, the user is automatically or proactively prompted to submit selection feedback. The user can express his or her satisfaction with the selection result through ratings, text evaluations, problem reports, etc. The collected feedback information is automatically classified by the intelligent selection platform, and positive evaluations are classified as positive feedback data sets, and negative evaluations or problem reports are classified as abnormal feedback data sets. According to the classified feedback information, it is stored in the positive feedback data set and the abnormal feedback data set (for example, feedback about the poor performance of a component in multiple production scenarios), respectively. Each piece of feedback information is accompanied by a timestamp and feedback source, which facilitates the intelligent selection platform to perform timing analysis.

[0061] Analyze the feedback classification data set, especially the abnormal feedback data set. By analyzing the time distribution and frequency of abnormal feedback data, identify which problems are concentrated in the short term and which are long-term, and adjust its own recommendation strategy through the timing compensation mechanism based on the analysis results. Based on the timing compensation mechanism, adjust its own recommendation algorithm and parameters. For example, when a large number of users are dissatisfied with the performance of a certain type of parts, the priority of such parts in future recommendations will be reduced, or the matching rules in the recommendation algorithm will be adjusted to avoid similar problems from happening again. At the same time, the information in the positive feedback data set will strengthen the recommendation logic corresponding to those satisfaction concentrations. In this way, the intelligent selection platform can continue to learn and improve. In the intelligent selection platform, by collecting user selection feedback, classifying the feedback, and self-optimizing the platform through the timing compensation mechanism, the intelligent selection platform analyzes user feedback and makes corresponding adjustments to make its own recommended parts selection more accurate and gradually improve its ability to match user needs.

[0062] Furthermore, the present application method also includes

[0063] Performing an abnormal cumulative evaluation on the abnormal feedback data set; if the abnormal cumulative evaluation result is higher than a preset threshold, generating a reset instruction; clearing the user's data on the intelligent selection platform according to the reset instruction, and executing the user reset.

[0064] Specifically, the abnormal feedback data set refers to a data set that collects and stores negative or problematic feedback from users on the recommendation results of the intelligent selection platform, usually indicating dissatisfaction with the component matching results of the intelligent selection platform; the abnormal cumulative evaluation is used to characterize the collected abnormal feedback data for cumulative analysis and evaluation, and to measure whether the user's negative feedback has accumulated to the extent that greater adjustments or resets are required; the preset threshold is used to measure whether the abnormal cumulative evaluation has reached the condition that requires intervention measures. When the accumulated abnormal feedback exceeds this value, it is considered that there is a serious problem with the user experience or matching results; the reset instruction refers to the instruction generated when the abnormal cumulative evaluation result is higher than the preset threshold, which clears the user's data and resets it; user reset means clearing the current user's account data, historical operation records, etc. after receiving the reset instruction, and restoring to the initial state to readjust the recommendation logic.

[0065] Execution steps: extract negative comments from user feedback information and store them in the abnormal feedback data set. Users may encounter problems during the selection process, and feedback such as recommended parts not meeting requirements, inaccurate selection results, etc., all of which are abnormal feedback; conduct concentration analysis on abnormal feedback data to evaluate whether the abnormal feedback submitted by users is too much or too concentrated within a period of time, calculate indicators such as the number of abnormal feedback, feedback frequency, and impact range, and give a cumulative score to reflect the severity of the problem.

[0066] The results of the abnormal cumulative evaluation are compared with the preset threshold. If the results of the cumulative evaluation are higher than the preset threshold, it means that the user's negative feedback has reached an unacceptable level. Furthermore, a reset instruction is automatically generated. The reset instruction is used to notify the intelligent selection platform to clear the data related to the user and start the reset process. After the reset instruction is triggered, the user's historical records, configuration data, etc. will be deleted, and the intelligent selection platform will be prepared to make intelligent selection recommendations for the user again; according to the generated reset instruction, the intelligent selection platform will clear the user's relevant data, including the user's operation records, selection preferences, historical matching call data, etc. on the intelligent selection platform, and let the user restart the entire selection process to avoid interference from previous abnormal feedback; after the reset, the selection process will be entered again, and the intelligent selection platform will recommend parts to the user based on the new input data and matching algorithm. At this time, the intelligent selection platform will clear the impact of previous incorrect matching after the reset operation, and improve the accuracy of future recommendations.

[0067] In summary, the beneficial effects of the embodiments of the present application are:

[0068] 1. Through the intelligent selection platform, data from different sources, such as design drawings, text and numerical information, can be integrated and automatically analyzed to improve data processing efficiency and accuracy.

[0069] 2. Through the fuzzy identification port and historical matching call function, when the user provides adjustment data or feedback, automatic compensation and correction are performed according to the priority database to ensure that the selection results meet user needs and enhance the flexibility and response speed of the system.

[0070] 3. By collecting user feedback and abnormal data, performing timing compensation and self-optimization, continuously improving the performance of the selection platform, establishing a user usage database and a general database, integrating historical matching data with current needs, improving the accuracy of selection and matching efficiency, and achieving personalized recommendations.

[0071] 4. As the abnormal feedback data set is evaluated by cumulative abnormality, if the result of the cumulative abnormality evaluation is higher than the preset threshold, a reset instruction is generated; according to the reset instruction, the user's data on the intelligent selection platform is cleared and the user reset is performed. After the reset operation, the intelligent selection platform will clear the influence of previous mismatches and improve the accuracy of future recommendations. Embodiment 2

[0072] Based on the same inventive concept as the method for intelligent selection of factory automation parts in the aforementioned embodiment, Figure 2 As shown, an embodiment of the present application provides a factory automation component intelligent selection device, wherein the device includes:

[0073] The port configuration module M100 is used to establish an intelligent selection platform, wherein the intelligent selection platform is configured with an input port, an adjustment port, and a fuzzy recognition port;

[0074] The data receiving module M200 is used to receive the user's input data through the input port, and the input data includes design drawings, project specifications, and production requirements. The input data is integrated and identified through the scenario establishment channel built into the intelligent selection platform, and the application scenario is output through the adjustment port. The application scenario is the result of the integrated identification;

[0075] The correction and compensation module M300 is used to read the user adjustment data of the adjustment port within a preset time window, and to correct and compensate the application scenario with the user adjustment data;

[0076] The data collection module M400 is used to activate the fuzzy recognition port and collect the user's fuzzy data, wherein the fuzzy data includes fuzzy input data and historical matching call data;

[0077] The intelligent selection module M500 is used to perform fuzzy analysis on the fuzzy input data in the fuzzy data, reconstruct the matching database with historical matching call data, use the fuzzy analysis results and the application scenarios after correction and compensation as matching data, perform component matching of the reconstructed matching database, and complete intelligent selection according to the component matching results.

[0078] Furthermore, the data receiving module M200 is also used to execute the following method:

[0079] Perform data analysis on the input data through the preprocessing layer and establish data analysis results;

[0080] Utilize the data analysis results to analyze the data types of the input data, and establish the first focus factor of the recognition layer according to the data type analysis results;

[0081] Use the data analysis results to conduct data richness analysis for each data type, and establish the second focus factor of the identification layer based on the data richness analysis results;

[0082] The recognition accuracy of the recognition layer is configured according to the first focus factor and the second focus factor, and the configured recognition layer is used to identify the data analysis results to establish an application scenario.

[0083] Furthermore, the data receiving module M200 is also used to execute the following method:

[0084] Synchronize the data analysis results corresponding to the design drawings to the image recognition layer, use the geometric shape features and material property features to extract data features, and establish the drawing analysis results;

[0085] Synchronize the data analysis results corresponding to the project manual to the natural language processing layer, use the natural language processing layer to extract key information from the text, and establish the text analysis results;

[0086] Synchronize the data analysis results corresponding to the production requirements to the numerical processing layer, use the numerical processing layer to digitize the numerical information, and establish a digital analysis result, wherein the image recognition layer, the natural language processing layer, and the numerical processing layer are all sub-processing units of the recognition layer, and the recognition accuracy is configured based on the first focus factor and the second focus factor;

[0087] Feature fusion is performed based on the drawing analysis results, text analysis results, and digital analysis results to complete the recognition.

[0088] Furthermore, the intelligent selection module M500 is also used to execute the following method:

[0089] Reading the user's account data and establishing the user's usage database based on the account data;

[0090] Parse the historical matching call data to obtain the user's single call interval, where the single call interval is the user's historical data usage preference interval;

[0091] A general database is established, and data of the general database and the usage database are fused based on a single call interval, and the data fusion result is used as the reconstructed matching database.

[0092] Furthermore, the correction and compensation module M300 is also used to perform the following method:

[0093] Identifying adjustment factors for the user adjustment data and establishing adjustment factor identification results;

[0094] Performing priority database matching based on the adjustment factor identification result, the priority database includes a first priority factor and a second priority factor, the first priority factor is an immediate compensation factor, and the second priority factor is a delayed compensation factor;

[0095] Complete application scenario correction and compensation based on priority database matching results.

[0096] Furthermore, the correction and compensation module M300 is also used to perform the following method:

[0097] Establish a balance adjustment weight constraint as follows:

[0098] ;

[0099] in, Adjust weights for users, Adjust the weights for the system, Adjust frequency for users, Characterizes the user-adjusted sensitivity coefficient, is the balance point, representing the standard user adjustment frequency, Normalized value for user-adjusted amplitude, is the influence coefficient of the adjustment range;

[0100] The user adjustment data is constrained by balancing the adjustment weight constraint to correct and compensate the application scenario.

[0101] Furthermore, the factory automation parts intelligent selection device is also used to perform the following method:

[0102] Obtaining user selection feedback, and classifying the selection feedback to establish a feedback classification data set, wherein the feedback classification data set includes a positive feedback data set and an abnormal feedback data set;

[0103] The feedback classification data set is used to establish timing compensation, and the platform self-optimization of the intelligent selection platform is performed through the timing compensation.

[0104] Furthermore, the factory automation parts intelligent selection device is also used to perform the following method:

[0105] Performing anomaly cumulative evaluation on the anomaly feedback data set;

[0106] If the abnormal cumulative evaluation result is higher than the preset threshold, a reset instruction is generated;

[0107] The user's data on the intelligent selection platform is cleared according to the reset instruction, and the user reset is performed.

[0108] In summary, any step can be stored as a computer instruction or program in an unlimited computer memory, and can be called and recognized by an unlimited computer processor, and no unnecessary restrictions are made here.

[0109] Furthermore, the above-mentioned technical scheme only reflects the preferred technical scheme of the technical scheme of the embodiment of the present application. Some changes that may be made to some parts thereof by technicians in this technical field all reflect the novel principles of the embodiment of the present application. Obviously, technicians in this field can make various changes and modifications to the present application without departing from the scope of the present application.

Claims

1. A method for intelligent selection of factory automation parts, characterized in that: The method comprises: Establishing an intelligent selection platform, wherein the intelligent selection platform is configured with an input port, an adjustment port, and a fuzzy recognition port; The input port is used to receive the user's input data, which includes design drawings, project specifications, and production requirements. The input data is integrated and identified through the scenario establishment channel built into the intelligent selection platform, and the application scenario is output through the adjustment port. The application scenario is the result of the integrated identification; Reading user adjustment data of the adjustment port within a preset time window, and correcting and compensating the application scenario with the user adjustment data; Activate the fuzzy recognition port to collect the user's fuzzy data, where the fuzzy data includes fuzzy input data and historical matching call data; After performing fuzzy analysis on the fuzzy input data in the fuzzy data, the matching database is reconstructed with the historical matching call data, and the fuzzy analysis results and the application scenarios after correction and compensation are used as matching data to perform component matching of the reconstructed matching database, and complete intelligent selection according to the component matching results; The reconstructing the matching database with the historical matching call data also includes: Reading the user's account data and establishing the user's usage database based on the account data; Parse the historical matching call data to obtain the user's single call interval, where the single call interval is the user's historical data usage preference interval; Establish a general database, fuse the general database and the usage database based on the single call interval, and use the data fusion result as the reconstructed matching database; The correcting and compensating the application scenario with the user adjustment data further includes: Identifying adjustment factors for the user adjustment data and establishing adjustment factor identification results; Performing priority database matching based on the adjustment factor identification result, the priority database includes a first priority factor and a second priority factor, the first priority factor is an immediate compensation factor, and the second priority factor is a delayed compensation factor; Complete application scenario correction and compensation based on priority database matching results; The method further comprises: Establish a balance adjustment weight constraint as follows: ； in, Adjust weights for users, Adjust the weights for the system, Adjust frequency for users, Characterizes the user-adjusted sensitivity coefficient, is the balance point, representing the standard user adjustment frequency, Normalized value for user-adjusted amplitude, is the influence coefficient of the adjustment range; The user adjustment data is constrained by balancing the adjustment weight constraint to correct and compensate the application scenario.

2. A method for intelligent selection of factory automation parts as claimed in claim 1, characterized in that: The integration and identification of input data through the scene establishment channel built into the intelligent selection platform also includes: Perform data analysis on the input data through the preprocessing layer and establish data analysis results; Utilize the data analysis results to analyze the data types of the input data, and establish the first focus factor of the recognition layer according to the data type analysis results; Use the data analysis results to conduct data richness analysis for each data type, and establish the second focus factor of the identification layer based on the data richness analysis results; The recognition accuracy of the recognition layer is configured according to the first focus factor and the second focus factor, and the configured recognition layer is used to identify the data analysis results to establish an application scenario.

3. A method for intelligent selection of factory automation parts as claimed in claim 2, characterized in that: The method of using the configured recognition layer to identify the data analysis result also includes: Synchronize the data analysis results corresponding to the design drawings to the image recognition layer, use the geometric shape features and material property features to extract data features, and establish the drawing analysis results; Synchronize the data analysis results corresponding to the project manual to the natural language processing layer, use the natural language processing layer to extract key information from the text, and establish the text analysis results; Synchronize the data analysis results corresponding to the production requirements to the numerical processing layer, use the numerical processing layer to digitize the numerical information, and establish a digital analysis result, wherein the image recognition layer, the natural language processing layer, and the numerical processing layer are all sub-processing units of the recognition layer, and the recognition accuracy is configured based on the first focus factor and the second focus factor; Feature fusion is performed based on the drawing analysis results, text analysis results, and digital analysis results to complete the recognition.

4. The method for intelligent selection of factory automation parts according to claim 1, characterized in that: The method further comprises: Obtaining user selection feedback, and classifying the selection feedback to establish a feedback classification data set, wherein the feedback classification data set includes a positive feedback data set and an abnormal feedback data set; The feedback classification data set is used to establish timing compensation, and the platform self-optimization of the intelligent selection platform is performed through the timing compensation.

5. A method for intelligent selection of factory automation parts as claimed in claim 4, characterized in that: The method further comprises: Performing anomaly cumulative evaluation on the anomaly feedback data set; If the abnormal cumulative evaluation result is higher than the preset threshold, a reset instruction is generated; The user's data on the intelligent selection platform is cleared according to the reset instruction, and the user reset is performed.

6. A factory automation parts intelligent selection device, characterized in that: A method for intelligently selecting factory automation parts according to any one of claims 1 to 5, the device comprising: A port configuration module is used to establish an intelligent selection platform, wherein the intelligent selection platform is configured with an input port, an adjustment port, and a fuzzy recognition port; The data receiving module is used to receive the user's input data through the input port, and the input data includes design drawings, project specifications, and production requirements. The input data is integrated and identified through the scenario establishment channel built into the intelligent selection platform, and the application scenario is output through the adjustment port. The application scenario is the result of the integrated identification; A correction and compensation module, used to read user adjustment data of the adjustment port within a preset time window, and to correct and compensate the application scenario with the user adjustment data; A data collection module, used to activate the fuzzy recognition port and collect the user's fuzzy data, wherein the fuzzy data includes fuzzy input data and historical matching call data; An intelligent selection module is used to perform fuzzy analysis on the fuzzy input data in the fuzzy data, reconstruct the matching database with historical matching call data, use the fuzzy analysis results and the application scenario after correction and compensation as matching data, perform component matching of the reconstructed matching database, and complete intelligent selection according to the component matching results; The reconstructing the matching database with the historical matching call data also includes: Reading the user's account data and establishing the user's usage database based on the account data; Parse the historical matching call data to obtain the user's single call interval, where the single call interval is the user's historical data usage preference interval; Establish a general database, fuse the general database and the usage database based on the single call interval, and use the data fusion result as the reconstructed matching database; The correcting and compensating the application scenario with the user adjustment data further includes: Identifying adjustment factors for the user adjustment data and establishing adjustment factor identification results; Performing priority database matching based on the adjustment factor identification result, the priority database includes a first priority factor and a second priority factor, the first priority factor is an immediate compensation factor, and the second priority factor is a delayed compensation factor; Complete application scenario correction and compensation based on priority database matching results; The device is also used to perform the following method, including: Establish a balance adjustment weight constraint as follows: ； in, Adjust weights for users, Adjust the weights for the system, Adjust frequency for users, Characterizes the user-adjusted sensitivity coefficient, is the balance point, representing the standard user adjustment frequency, Normalized value for user-adjusted amplitude, is the influence coefficient of the adjustment range; The user adjustment data is constrained by balancing the adjustment weight constraint to correct and compensate the application scenario.

Images