Screw machine control management method and system, electronic equipment and storage medium

Abstract

The invention discloses a screw machine control management method and system, electronic equipment and a storage medium, and relates to the field of screw machines. The method comprises the following steps: acquiring a target image, target sensor data and target sound; integrating the target image, the target sensor data and the target sound to obtain target data, and processing the target data through a preset deep learning model to obtain a diagnosis result; operating parameters of the screw machine are adjusted according to the diagnosis result and the target sensor data, the production result of the adjusted screw machine is obtained, and the operating parameters comprise the output power of a screwdriver, the tightening speed and the positioning precision of the screw machine; and determining a preset production result according to the inventory data and actual demands, comparing the production result with the preset production result to obtain a comparison result, and adjusting the operation parameters according to the comparison result. By implementing the technical scheme provided by the invention, the working efficiency and the product quality of the screw machine are improved, the failure rate is reduced, and accurate control and intelligent management are realized.

Description

Technical Field

[0001] This application relates to the technical field of screwdrivers, and specifically relates to a control and management method, system, electronic device, and storage medium for a screwdriver. Background Art

[0002] A screwdriver is an automated device widely used in modern manufacturing, mainly for realizing the automatic assembly of fasteners such as screws and bolts. With the progress of industrial technology, a screwdriver not only needs to be efficient and accurate, but also needs to be able to monitor and optimize the production process in real time to improve product quality and reduce failure rates.

[0003] Regarding the problem of quality monitoring, the traditional approach relies on workers to regularly and randomly select a certain proportion of product samples for appearance inspection. This method is labor-intensive and material-consuming and does not have the characteristics of continuity and full coverage; or by using specialized tools designed to measure aspects such as torque magnitude and position accuracy to evaluate whether the attribute characteristics of each batch of finished products meet the standard specified value range. Although this method can accurately quantify various parameter indicators, it cannot prevent unqualified products from flowing into the next process because it can only be verified afterwards, and it also faces the problems of long cycle and slow response.

[0004] Therefore, a control and management method for a screwdriver that improves the accuracy and efficiency of quality monitoring is needed. Summary of the Invention

[0005] This application provides a control and management method, system, electronic device, and storage medium for a screwdriver, which realizes the refined control and management of the screwdriver through key technical links such as multi-modal data acquisition, deep learning diagnosis, intelligent parameter adjustment, and production result comparison.

[0006] In the first aspect of this application, a control and management method for a screwdriver is provided, which is applied to a screwdriver. The method includes: Obtain a target image, target sensor data, and target sound. The target image includes the installation position and posture of the screw and the image during the screw tightening process. The target sensor data includes the torque, displacement, and temperature data during the screw tightening process. The target sound includes the sound generated by the screwdriver during operation; Integrate the target image, the target sensor data, and the target sound to obtain target data, and process the target data through a preset deep learning model to obtain a diagnosis result; Adjust the operating parameters of the screwdriver according to the diagnosis result and the target sensor data, and obtain the production result of the adjusted screwdriver. The operating parameters include the output power of the screwdriver, the tightening speed, and the positioning accuracy of the screwdriver; Determine the preset production result according to the inventory data and the actual demand, compare the production result with the preset production result to obtain a comparison result, and adjust the operating parameters according to the comparison result.

[0007] Optionally, the integrating the target image, the target sensor data, and the target sound to obtain target data includes: Preprocess the target image, the target sensor data, and the target sound, and synchronize the target image, the target sensor data, and the target sound in time by interpolation; Extract visual features from the preprocessed target image, where the visual features include edge features and texture features, extract physical features from the preprocessed target sensor data, where the physical features include torque mean and displacement variance, and extract audio features from the preprocessed target sound, where the audio features include spectral features and time-domain features; Fuse the visual features, the physical features, and the audio features to construct a comprehensive feature vector, and use the comprehensive feature vector as the target data.

[0008] Optionally, the processing the target data by a preset deep learning model to obtain a diagnosis result includes: Combine the visual features and the physical features, and analyze the first correlation between the appearance quality and the tightening effect of the screw through the connection of neurons and the non-linear transformation of the activation function; Combine the visual features and the audio features, and analyze the second correlation between the appearance state and the running sound of the screw; Combine the physical features and the audio features, and analyze the third correlation between the physical changes and the sound performance during the screw tightening process through the calculation of the hidden layer; Determine the diagnosis result according to the first correlation, the second correlation, and the third correlation.

[0009] Optionally, the determining the diagnosis result according to the first correlation, the second correlation, and the third correlation includes: Perform weighted processing on the first correlation, the second correlation, and the third correlation according to a preset weight allocation mechanism to obtain a comprehensive diagnosis score; If the comprehensive diagnosis score is greater than or equal to a preset threshold, output a diagnosis result that the screw tightening is normal; If the comprehensive diagnosis score is less than the preset threshold, output the fault type according to the comprehensive diagnosis score.

[0010] Optionally, the adjusting the operating parameters of the screwdriver according to the diagnosis result and the target sensor data includes: Determine the target operating parameters corresponding to the diagnosis result according to the relationship between the fault type and the operating parameters, and determine the tightening quality influence coefficient and the adjustment difficulty coefficient of each target operating parameter; Perform weighted processing according to the tightening quality influence coefficient and the adjustment difficulty coefficient to obtain the adjustment value of each target operating parameter, and determine the adjustment priority of the multiple target operating parameters according to the adjustment value; Adjust the multiple target operating parameters according to the adjustment priority, and monitor the change of the target sensor data in real time.

[0011] Optionally, the determining the preset production result according to the inventory data and the actual demand includes: Obtain the inventory quantity and inventory turnover rate of the target parts, and predict the inventory demand in a future preset time period through the inventory quantity and the inventory turnover rate; Obtain the urgency and quantity requirements in the customer order information to determine the actual demand; Determine the preset production result through the inventory demand and the actual demand, and the preset production result includes the production beat, the output target and the product quality standard.

[0012] Optionally, the comparing the production result with the preset production result to obtain a comparison result, and adjusting the operating parameters according to the comparison result includes: Compare the key indicators in the production result with the corresponding indicators in the preset production result one by one to determine the deviation type and deviation degree, and the key indicators include the output and the product quality qualification rate; Match the optimal operating parameter adjustment scheme from the preset parameter adjustment strategy library according to the deviation type and the deviation degree.

[0013] In the second aspect of the present application, a screwdriver control and management system is provided, including an acquisition module, an integration module, an adjustment module and an execution module, wherein: The acquisition module is configured to obtain a target image, target sensor data and target sound. The target image includes the installation position and posture of the screw and the image during the screw tightening process. The target sensor data includes the torque, displacement and temperature data during the screw tightening process. The target sound includes the sound generated by the screwdriver during operation; The integration module is configured to integrate the target image, the target sensor data and the target sound to obtain target data, and process the target data through a preset deep learning model to obtain a diagnosis result; An adjustment module, configured to adjust the operating parameters of the screwdriver machine according to the diagnostic result and the target sensor data, and obtain the production result of the adjusted screwdriver machine. The operating parameters include the output power of the screwdriver, the tightening speed, and the positioning accuracy of the screwdriver machine. An execution module, configured to determine a preset production result according to the inventory data and the actual demand, compare the production result with the preset production result to obtain a comparison result, and adjust the operating parameters according to the comparison result.

[0014] In a third aspect of the present application, an electronic device is provided, including a processor, a memory, a user interface, and a network interface. The memory is used to store instructions. Both the user interface and the network interface are used to communicate with other devices. The processor is used to execute the instructions stored in the memory, so that the electronic device executes the method described in any one of the above.

[0015] In a fourth aspect of the present application, a computer-readable storage medium is provided. The computer-readable storage medium stores instructions, and when the instructions are executed, the method described in any one of the above is executed.

[0016] In summary, one or more technical solutions provided in the embodiments of the present application have at least the following technical effects or advantages: 1. By performing real-time processing on the integrated target data through a deep learning model, various faults that occur during the screw tightening process of the screwdriver machine can be quickly and accurately diagnosed, such as screw position deviation, insufficient tightening torque, etc. Once a problem is found, the system immediately adjusts the operating parameters according to the diagnostic result and the target sensor data, eliminates the fault in time, avoids long-term downtime and waiting during the production process, and significantly improves production efficiency; 2. Based on the preset production result determined according to the inventory data and the actual demand, and the comparison and analysis of the production result with the preset target, the system can dynamically adjust the production rhythm of the screwdriver machine. When the market demand is strong and the inventory is tight, the production rhythm is accelerated to increase the output; when the market demand is stable and the inventory is sufficient, the production rhythm is appropriately reduced to balance production and inventory, avoid waste of resources, and make the production process more efficiently and flexibly adapt to market changes; 3. The intelligent adjustment of operating parameters, such as the output power of the screwdriver, the tightening speed, and the positioning accuracy of the screwdriver machine, can ensure that each tightening operation meets the expected quality standards. For example, by adjusting the output power and the tightening speed, the tightening torque and depth of the screw meet the product requirements, avoiding screw damage or loose fastening caused by too large or too small torque; improving the positioning accuracy to ensure that the screw is accurately aligned with the screw hole, reducing skew and slipping phenomena, thereby significantly improving the tightening quality of the screw; 4. The integration and analysis of multi-modal data enable the system to comprehensively monitor the tightening quality of screws from multiple dimensions. The target image provides information about the appearance of the screws, the target sensor data reflects the changes in physical parameters during the tightening process, and the target sound captures abnormal sounds during the operation of the screwdriver. By integrating this information, the system can more accurately identify potential quality problems, such as scratches on the screw surface and missing screw heads, and take timely measures to handle them, ensuring the stability and consistency of product quality; 5. The system collects and integrates a large amount of production data, including target images, target sensor data, target sounds, and production results. Through the analysis and mining of these data, it provides scientific and accurate decision-making basis for production managers. Managers can optimize the production process, adjust resource allocation, and formulate reasonable production plans based on data feedback, improving the scientificity and effectiveness of management. Description of the Drawings

[0017] Figure 1 is a schematic flowchart of the screw machine control and management method disclosed in the embodiments of the present application; Figure 2 is a schematic diagram of the modules of the screw machine control and management system disclosed in the embodiments of the present application; Figure 3 is a schematic structural diagram of an electronic device disclosed in the embodiments of the present application.

[0018] Description of the Reference Numerals: 201, acquisition module; 202, integration module; 203, adjustment module; 204, execution module; 301, processor; 302, communication bus; 303, user interface; 304, network interface; 305, memory. Detailed Embodiments

[0019] In order to enable those skilled in the art to better understand the technical solutions in this specification, the technical solutions in the embodiments of this specification will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of this specification. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments.

[0020] In the description of the embodiments of the present application, words such as "for example" or "for illustration" are used to indicate examples, illustrations, or explanations. Any embodiment or design solution described as "for example" or "for illustration" in the embodiments of the present application should not be construed as being more preferred or having more advantages than other embodiments or design solutions. Rather, the use of words such as "for example" or "for illustration" is intended to present relevant concepts in a specific manner.

[0021] In the description of the embodiments of the present application, the term "plurality" means two or more. For example, a plurality of systems means two or more systems, and a plurality of screen terminals means two or more screen terminals. In addition, the terms "first" and "second" are only used for descriptive purposes and cannot be construed as indicating or implying relative importance or implicitly indicating the technical features indicated. Thus, the features defined with "first" and "second" may explicitly or implicitly include one or more of such features. The terms "comprise", "include", "have" and their variants all mean "including but not limited to", unless otherwise specifically emphasized in other ways.

[0022] This embodiment discloses a method for controlling and managing a screwdriver machine, which is applied to the screwdriver machine. Figure 1 It is a schematic flowchart of the method for controlling and managing the screwdriver machine disclosed in the embodiments of the present application. As Figure 1 shown, the method includes the following steps: S101. Obtain a target image, target sensor data, and target sound. The target image includes the installation position and attitude of the screw and the image during the screw tightening process. The target sensor data includes the torque, displacement, and temperature data during the screw tightening process. The target sound includes the sound generated by the screwdriver machine during operation; S102. Integrate the target image, the target sensor data, and the target sound to obtain target data, and process the target data through a preset deep learning model to obtain a diagnosis result; S103. Adjust the operating parameters of the screwdriver machine according to the diagnosis result and the target sensor data, and obtain the production result of the adjusted screwdriver machine. The operating parameters include the output power of the screwdriver, the tightening speed, and the positioning accuracy of the screwdriver machine; S104. Determine a preset production result according to the inventory data and the actual demand, compare the production result with the preset production result to obtain a comparison result, and adjust the operating parameters according to the comparison result.

[0023] Use a high-resolution industrial camera to take real-time pictures of the installation position, posture, and tightening process of the screws to obtain clear target images. The target images can intuitively show the specific situation of the screws during the assembly process, such as whether the screws are accurately aligned with the screw holes and whether the screw heads are intact. Install a variety of intelligent sensors at key parts of the screwdriver, such as high-precision torque sensors, displacement sensors, and temperature sensors, etc., and obtain target sensor data from the intelligent sensors. The torque sensor monitors the torque change during the screw tightening process in real time, the displacement sensor records the displacement trajectory of the screw, and the temperature sensor detects the temperature fluctuation when the screwdriver is running. The target sensor data provides accurate quantitative indicators for evaluating the tightening quality and equipment status. Use a professional sound acquisition device, such as a microphone array, to record the target sound generated by the screwdriver during operation. The target sound contains rich information, such as the normal operation sound when the screw is tightened, abnormal noises when the screw is stripped or the equipment fails, etc. By analyzing the sound signal, it can assist in judging the running state and tightening quality of the screwdriver. Preprocess the obtained target images, target sensor data, and target sounds, such as image denoising, sensor data filtering, sound noise reduction, etc., to improve the data quality. Then, synchronize the data of different modalities in time through methods such as interpolation, and fuse them into a comprehensive target dataset. The target dataset comprehensively reflects various information of the screwdriver during the screw tightening process. Input the integrated target data into a preset deep learning model. The model has been trained with a large amount of screwdriver operation data and has powerful feature extraction and pattern recognition capabilities. It can analyze the correlations between the appearance quality and tightening effect of the screws, the correlation between the appearance state and running sound, the correlation between physical changes and sound performance, etc. from the target data. Finally, the model outputs diagnostic results, such as normal screw tightening, screw position deviation, insufficient tightening torque, etc. According to the diagnostic results and target sensor data, the system automatically adjusts the operating parameters of the screwdriver. For example, if the diagnostic result shows a screw position deviation, the system will increase the positioning accuracy parameters of the screwdriver, such as adjusting the focus distance of the camera and optimizing the positioning algorithm; if the tightening torque is insufficient, it will increase the output power of the screwdriver or adjust the tightening speed. The adjusted operating parameters include the output power of the screwdriver, tightening speed, positioning accuracy of the screwdriver, etc. After adjusting the operating parameters, the screwdriver continues to run, and the system monitors the adjusted production results in real time. The production results include key indicators such as the tightening quality of the screws (such as whether the torque meets the requirements and whether the screw position is accurate), production efficiency (such as the number of screws tightened per unit time), etc. By comparing the production results before and after adjustment, evaluate the actual effect of the operating parameter adjustment. Determine the preset production results of the screwdriver according to the inventory data and actual demand. The inventory data provides information on the current inventory quantity of screws and related components, and the actual demand includes customer order quantity, market trends, etc. Considering these factors comprehensively, set the preset production results, such as production rhythm, output target, product quality standard, etc.Compare the actual production results of the screwdriver machine with the preset production results, and calculate the deviation values or completion percentage of various indicators. For example, if the actual output is lower than the preset output, calculate the shortfall in output; if the qualified rate of product quality is lower than the preset standard, calculate the proportion of unqualified products. According to the comparison results, adjust the operating parameters of the screwdriver machine again. If the production results do not meet the preset goals, analyze the reasons for the deviation, and it may be necessary to further optimize the operating parameters. For example, if the output is insufficient, it may be necessary to further increase the output power and tightening speed of the screwdriver; if the product quality problem is prominent, it is necessary to more finely adjust parameters such as positioning accuracy and clamping force to achieve the preset production goals.

[0024] By integrating the target image, target sensor data, and target sound, comprehensive monitoring of the screw tightening process is achieved. The fusion of this multi-dimensional data enables the diagnostic system to more accurately capture abnormal situations during the screw tightening process, improving the accuracy and reliability of diagnosis. Using a preset deep learning model to process the integrated target data can automatically identify potential problems during the screw tightening process, such as incorrect screw installation positions, abnormal postures, insufficient or excessive torque, abnormal displacement, too high temperature, and abnormal operating sounds of the screwdriver machine. This intelligent diagnostic method greatly reduces manual intervention and improves production efficiency and accuracy. According to the diagnostic results and target sensor data, the system can automatically adjust the operating parameters of the screwdriver machine, such as the output power of the screwdriver, tightening speed, and positioning accuracy of the screwdriver machine. This adaptive adjustment method enables the screwdriver machine to flexibly adjust according to the actual situation, ensuring the consistency and reliability of screw tightening. By comparing the production results with the preset production results, the system can promptly detect deviations in the production process and further adjust the operating parameters according to the comparison results. This continuous optimization process helps improve production efficiency, reduce waste, and ensure that product quality meets the preset standards. The entire process realizes intelligent management from data collection, processing, diagnosis to parameter adjustment, improving the intelligent level of the production line. This intelligent management method helps reduce production costs, improve production efficiency, and provide strong support for the digital transformation of enterprises.

[0025] Optionally, the integrating the target image, the target sensor data, and the target sound to obtain target data includes: Preprocess the target image, the target sensor data, and the target sound, and synchronize the target image, the target sensor data, and the target sound in time through interpolation; Extract visual features from the preprocessed target image, where the visual features include edge features and texture features, extract physical features from the preprocessed target sensor data, where the physical features include torque mean and displacement variance, and extract audio features from the preprocessed target sound, where the audio features include spectral features and time-domain features; Fuse the visual features, the physical features, and the audio features to construct a comprehensive feature vector, and use the comprehensive feature vector as the target data.

[0026] Denoise the acquired target image, using algorithms such as Gaussian filtering to eliminate the noise in the image and make the image clearer; then perform grayscale processing to convert the color image into a grayscale image and simplify the processing complexity of the image data; finally, perform normalization processing to scale the image pixel values to a unified range (such as between 0 and 1) for subsequent feature extraction. Filter the target sensor data, using algorithms such as low-pass filters to remove the high-frequency noise in the data and retain the useful signals; then perform interpolation processing to fill in the missing values in the data and ensure the continuity of the data; then perform unit unification processing to convert the data of different sensors into the same unit for convenient subsequent feature extraction. Reduce the noise of the target sound, using algorithms such as spectral subtraction to eliminate the background noise and retain the useful components in the sound signal; then perform echo cancellation processing to eliminate the echo interference in the sound signal and make the sound signal purer. Through the interpolation algorithm, align the preprocessed target image, target sensor data, and target sound in time. For example, if the acquisition frequency of the target image is different from the sampling frequency of the target sensor data, methods such as linear interpolation or spline interpolation can be used to interpolate the image data and sensor data to the same time point to ensure that in the subsequent feature extraction and fusion processes, the data of different modalities can correspond to the operating state of the screwdriver at the same moment. Extract visual features from the preprocessed target image, mainly including edge features and texture features. Edge features are extracted through edge detection algorithms (such as the Canny algorithm), which can outline the contour and shape of the screw and reflect the installation position and attitude information of the screw; texture features are extracted through algorithms such as the gray-level co-occurrence matrix (GLCM), which can describe the roughness and texture pattern of the screw surface and provide a basis for identifying the appearance quality of the screw. Extract physical features from the preprocessed target sensor data, mainly including the mean torque and displacement variance. The mean torque is obtained by calculating the average value of the torque data within a certain time window, which can reflect the average torque magnitude during the screw tightening process and is an important indicator for evaluating the tightening quality; the displacement variance is obtained by calculating the variance of the displacement data, which can reflect the position stability of the screw during the tightening process. The smaller the displacement variance, the more stable the tightening position of the screw. Extract audio features from the preprocessed target sound, mainly including spectral features and time-domain features. Spectral features convert the sound signal from the time domain to the frequency domain through algorithms such as Fourier transform to extract the frequency components of the sound signal. Abnormal frequency components are often associated with faults in the screwdriver or abnormal tightening states of the screw; time-domain features include the duration and intensity change of the sound signal, which can reflect the dynamic characteristics of the sound during the operation of the screwdriver. Concatenate the extracted visual features, physical features, and audio features to construct a comprehensive feature vector. Specifically, connect the feature vectors such as edge features, texture features, mean torque, displacement variance, spectral features, and time-domain features in sequence to form a comprehensive feature vector containing all feature information.On the basis of feature splicing, the features of different modalities are weighted and fused. Different weights are assigned to them according to the importance and reliability of each modality feature in the fault diagnosis of the screwdriver machine. For example, if the visual feature has high accuracy in identifying the screw installation position, a larger weight is given; if the physical feature plays a key role in evaluating the tightening quality, a larger weight is also given. Through weighted fusion, key features can be highlighted, and the diagnostic ability of the comprehensive feature vector can be improved. The fused comprehensive feature vector is used as the target data and input into a preset deep learning model for processing. The target data contains rich multi-modal information during the operation of the screwdriver machine, which can comprehensively reflect the installation state of the screw, the tightening quality, and the operation status of the equipment, providing a solid data basis for the diagnostic analysis of the deep learning model.

[0027] By preprocessing the target image, target sensor data, and target sound, and synchronizing them in time through interpolation, the alignment of different modality data in the time dimension is ensured. This synchronous processing eliminates the time deviation that may occur during the data acquisition process, enabling subsequent data analysis and fusion to be based on the complete information at the same time point, improving the consistency and accuracy of the data, and laying a foundation for accurately diagnosing the operation status of the screwdriver machine. Visual features, including edge features and texture features, are extracted from the preprocessed target image, which can capture key information such as the appearance and surface condition of the screw. The edge features clearly outline the contour of the screw, and the texture features reflect the roughness and detail changes of the screw surface, providing an intuitive basis for evaluating the installation position and posture of the screw. Physical features, such as the mean torque and displacement variance, are extracted from the target sensor data to quantify the key physical quantities during the screw tightening process. The mean torque reflects the average tightening force of the screw, and the displacement variance reveals the stability of the screw position change during the tightening process, providing accurate data support for analyzing the tightening quality of the screw and the operation stability of the equipment. Audio features, including spectral features and time-domain features, are extracted from the target sound. The spectral features can reflect the frequency distribution of the sound signal during the operation of the screwdriver machine, and the time-domain features describe the temporal characteristics of the sound signal, such as the duration and intensity change of the sound. These audio features provide important clues for identifying possible abnormal vibrations and noises during the screw tightening process. The visual features, physical features, and audio features are fused to construct a comprehensive feature vector. This fusion method enables the model to comprehensively understand the operation status of the screwdriver machine from multiple dimensions, considering multiple factors such as the appearance of the screw, physical tightening parameters, and operation sound. For example, when diagnosing the screw tightening quality, not only the visual features are used to judge whether the installation position of the screw is accurate, but also the physical features are combined to analyze whether the tightening torque meets the requirements, and the audio features are referred to judge whether there are abnormal sounds during the tightening process, so as to obtain a more accurate and comprehensive diagnostic result.

[0028] Optionally, the processing of the target data by the preset deep learning model to obtain a diagnosis result includes: Combining visual features with physical features, and through the connection of neurons and the non-linear transformation of activation functions, analyzing the first correlation between the appearance quality and the tightening effect of the screw; Combining visual features with audio features, and analyzing the second correlation between the appearance state and the running sound of the screw; Combining physical features with audio features, and through the calculation of the hidden layer, analyzing the third correlation between the physical changes and the sound performance during the screw tightening process; Determining the diagnosis result according to the first correlation, the second correlation, and the third correlation.

[0029] The comprehensive feature vector is input into the input layer of the deep learning model, and each eigenvalue corresponds to an input node. The model performs preliminary processing on the input feature vector, including operations such as feature normalization, to adapt to subsequent neural network calculations. The comprehensive feature vector enters multiple hidden layers of the model for calculation. Each hidden layer consists of several neurons, and the neurons are connected by weights. In the hidden layer, the model uses activation functions (such as ReLU, sigmoid, etc.) to perform non-linear transformations on the input features, extract deep feature representations, and capture complex associations and interaction information between different modality data. The model combines visual features (edge features and texture features) with physical features (torque mean and displacement variance), and through the complex connections of neurons and the non-linear transformations of activation functions, analyzes the association between the appearance quality and tightening effect of the screw. For example, the edge feature shows the clarity and integrity of the screw head edge. Combined with the torque mean, it can be judged whether the insufficient tightening force is caused by appearance defects (such as screw head deformation) during the tightening process of the screw; the texture feature reflects the roughness of the screw surface. Combined with the displacement variance, it can be analyzed whether the screw position is unstable due to the uneven surface during tightening. In actual production, if there are scratches or unevenness on the screw surface, it may affect the friction and tightening effect during tightening. Through the first correlation analysis, the model can identify the relationship between this appearance quality and tightening effect, providing a basis for optimizing the screw appearance design and improving the tightening quality. The model combines visual features with audio features to analyze the association between the appearance state and running sound of the screw. For example, the edge feature shows whether the installation position of the screw is accurate. Combined with the spectral feature, it can be judged whether abnormal friction sounds are generated due to position deviation during the tightening process of the screw; the texture feature combined with the time-domain feature can analyze the influence of the screw surface condition on the running sound stability. During the operation of the screw machine, if the screw installation position is incorrect, it may cause abnormal noise in the equipment. Through the second correlation analysis, the model can capture the relationship between this appearance state and sound, and timely detect problems with the screw installation position, avoiding equipment damage or poor screw tightening caused by position deviation. The model combines physical features with audio features, and through the calculation of the hidden layer, analyzes the association between the physical changes and sound performance during the screw tightening process. For example, the torque mean combined with the spectral feature can judge whether the change in the tightening torque causes a change in the frequency components of the equipment running sound; the displacement variance combined with the time-domain feature can analyze the relationship between the stability of the screw position change and the fluctuation of the running sound intensity during the tightening process. When tightening the screw, if the torque suddenly increases, it may cause a sharp noise in the equipment. Through the third correlation analysis, the model can identify the relationship between this physical change and sound performance, providing important information for monitoring the operation state of the equipment and preventing potential failures. According to the analysis results of the first, second, and third correlations, the model comprehensively evaluates the operation state and tightening quality of the screw machine.For example, if the first association shows that the appearance quality of the screw is good and matches the tightening effect, the second association does not find an association between the appearance state and abnormal sound, and the third association also does not detect an abnormal relationship between physical changes and sound performance, it can be determined that the screw machine is operating normally and the tightening quality meets the requirements. If an abnormality is found in a certain association analysis, the model will determine the corresponding fault type based on the specific association relationship and abnormal characteristics. For example, if the first association finds that the appearance defect of the screw is associated with insufficient tightening torque, it is diagnosed as "fault of insufficient tightening torque caused by screw appearance quality"; if the third association detects an association between torque change and abnormal noise, it is diagnosed as "tightening quality problem caused by abnormal equipment operation". Finally, the model outputs the diagnostic result as a specific label or probability value, such as "normal", "screw appearance defect", "abnormal equipment operation", etc., providing an accurate guiding basis for the maintenance and adjustment of the screw machine.

[0030] By combining visual features, physical features, and audio features, deep fusion of multi-modal data is achieved. This fusion method can comprehensively analyze the operating state of the screw machine from multiple dimensions, considering various factors such as the appearance quality of the screw, tightening effect, and operating sound. For example, when analyzing the association between the appearance quality of the screw and the tightening effect, not only is the installation position of the screw judged based on visual features, but also data such as the average torque in physical features is combined to determine whether the tightening torque meets the requirements, thereby obtaining a more accurate diagnostic result. Through the complex connection of neurons and the non-linear transformation of activation functions, the deep learning model can capture the complex association relationships between different modal features. For example, when analyzing the association between physical changes and sound performance, the model can identify the subtle connection between torque change and sound frequency, and find that abnormal torque fluctuations may lead to specific sound abnormalities, thus more accurately diagnosing potential faults of the screw machine. The features of different modalities are complementary and can complement each other's information, improving the model's adaptability to various situations. For example, in some cases, a single visual feature may not be able to accurately judge the tightening quality of the screw, but by combining the change of sound signal in audio features, the model can more comprehensively evaluate abnormal situations during the tightening process, enhancing the model's generalization ability under different working conditions. The multi-modal feature fusion makes the model more robust in the face of noise and incomplete data. Even if the data of a certain modality is noisy or missing, the data of other modalities can still provide effective supplementary information to help the model accurately diagnose the operating state of the screw machine.

[0031] Optionally, the determining the diagnostic result according to the first association, the second association, and the third association includes: Performing weighted processing on the first association, the second association, and the third association according to a preset weight allocation mechanism to obtain a comprehensive diagnostic score; If the comprehensive diagnostic score is greater than or equal to a preset threshold, outputting a diagnostic result indicating that the screw is tightened normally; If the comprehensive diagnosis score is less than the preset threshold, the fault type is output according to the comprehensive diagnosis score.

[0032] The preset weighting mechanism is based on the importance and reliability of each correlation analysis in screw machine fault diagnosis. For example, if the correlation between the screw's appearance and tightening effect (the first correlation) has the greatest impact on tightening quality and high diagnostic accuracy, it is assigned a higher weight. If the correlation between the screw's appearance and operating sound (the second correlation), while also providing some diagnostic information, is less important than the first correlation, it is assigned a lower weight. The correlation between physical features and audio features (the third correlation) is weighted based on its contribution to the diagnostic result. To determine the diagnostic result, the analysis results of the first, second, and third correlations are multiplied by the corresponding weight coefficients. The weighted results are then summed or combined to produce a comprehensive diagnostic score. This comprehensive diagnostic score integrates all information from the multimodal feature fusion analysis and provides a more comprehensive reflection of the screw machine's operating status and tightening quality. The preset threshold is set based on the screw machine's normal operating status and tightening quality standards. It is a critical value used to distinguish whether a screw is tightened normally. Thresholds must be set based on historical data and expert experience to ensure that the comprehensive diagnostic score remains consistently above or equal to the threshold under normal conditions, while falling below it in the event of a fault. When the comprehensive diagnostic score is greater than or equal to the preset threshold, the screw machine is operating properly and the tightening quality meets requirements. In this case, the system outputs a diagnostic result of "Normal Screw Tightening," indicating that the production process can proceed smoothly without additional intervention, which helps improve production efficiency and equipment utilization. When the comprehensive diagnostic score is less than the preset threshold, it indicates that the screw machine may have a fault or anomaly during the tightening process. Based on the specific value of the comprehensive diagnostic score and detailed information from each correlation analysis, the system outputs the specific fault type. For example, if the score for the first correlation is low, the diagnosis may be "screw installation position deviation"; if the score for the third correlation is abnormal, the diagnosis may be "unstable torque during screw tightening, resulting in abnormal sound." These fault diagnosis results provide clear guidance and basis for subsequent troubleshooting and resolution.

[0033] By performing weighted processing on the first correlation between the appearance quality and tightening effect of the screw, the second correlation between the appearance state and running sound, and the third correlation between physical changes and sound performance, the complex relationships among multi-modal data are comprehensively evaluated. This comprehensive evaluation method can more comprehensively reflect the actual operating state of the screw machine, avoid the one-sidedness that may be brought by single-modal correlation analysis, and thus improve the accuracy of the diagnostic results. The preset weight allocation mechanism is reasonably allocated according to the importance and contribution degree of each correlation in the screw machine fault diagnosis. This optimization of weight allocation makes the diagnostic results more in line with the fault characteristics and influencing factors in actual production, further improving the accuracy of diagnosis. The weighted results of the first correlation, the second correlation, and the third correlation are converted into a comprehensive diagnostic score, providing a quantitative indicator for fault identification. The comprehensive diagnostic score can intuitively reflect the deviation degree between the operating state of the screw machine and the normal standard, simplify the process of fault identification, and enable the system to quickly determine whether there is a fault in the screw machine. The diagnostic results are obtained based on clear correlation analysis, including the specific correlations between appearance quality and tightening effect, appearance state and running sound, and physical changes and sound performance. This clear correlation analysis provides a clear logical basis for the diagnostic results, enabling operators and maintenance engineers to better understand the causes and influencing factors of faults, enhancing the interpretability of the diagnostic results, and contributing to subsequent fault handling and equipment maintenance. When the comprehensive diagnostic score is lower than the threshold, the system can further output specific fault types according to the score, such as "screw position deviation", "insufficient tightening torque", etc. This accurate fault type positioning provides a clear fault handling direction for maintenance personnel, reduces the difficulty and complexity of fault troubleshooting, and improves the maintenance efficiency and accuracy.

[0034] Optionally, the adjusting the operating parameters of the screw machine according to the diagnostic result and the target sensor data includes: Determining the target operating parameters corresponding to the diagnostic result according to the relationship between the fault type and the operating parameters, and determining the tightening quality influence coefficient and the adjustment difficulty coefficient of each target operating parameter; Performing weighted processing according to the tightening quality influence coefficient and the adjustment difficulty coefficient to obtain the adjustment value of each target operating parameter, and determining the adjustment priority of the multiple target operating parameters according to the adjustment value; Adjusting the multiple target operating parameters according to the adjustment priority, and real-time monitoring the change of the target sensor data.

[0035] Based on the fault types in the diagnostic results, such as "screw position deviation", "insufficient tightening torque", etc., combined with the working principle and empirical knowledge of the screwdriver machine, determine the operating parameters directly related to these fault types. For example, "screw position deviation" may be closely related to the positioning accuracy parameter of the screwdriver machine, while "insufficient tightening torque" may be related to the output power parameter of the screwdriver. For each target operating parameter, evaluate its impact degree on the screw tightening quality, that is, the tightening quality impact coefficient. This coefficient reflects the sensitivity of parameter adjustment to the final tightening quality. For example, the output power of the screwdriver may have a very significant impact on the tightening torque, so its tightening quality impact coefficient is relatively high; while some parameters may also affect the tightening quality, but the impact is relatively small, and their coefficients are relatively low. At the same time, determine the adjustment difficulty coefficient of each target operating parameter, which reflects the difficulty and required resources for adjusting this parameter. Some parameters may be difficult to adjust due to factors such as equipment structure and control system, requiring more time, manpower or technical input, and their adjustment difficulty coefficients are relatively high; while other parameters are relatively easy to adjust, and their coefficients are relatively low. According to the tightening quality impact coefficient and the adjustment difficulty coefficient, perform weighted processing on each target operating parameter to calculate its adjustment value. The adjustment value comprehensively considers the potential for improving the tightening quality by parameter adjustment and the feasibility of actual adjustment, providing a quantitative basis for subsequent parameter adjustment. For example, for a parameter with a high tightening quality impact coefficient and a low adjustment difficulty coefficient, its adjustment value may be relatively large, indicating that preferentially adjusting this parameter can effectively improve the tightening quality and the adjustment process is relatively easy. According to the calculated adjustment values, sort multiple target operating parameters to determine their adjustment priorities. Parameters with higher priorities will be adjusted first to quickly solve the problems that have the greatest impact on the tightening quality and are easy to adjust, thereby improving production efficiency and product quality. For example, in the face of multiple fault types, if both the "output power of the screwdriver" and the "positioning accuracy of the screwdriver machine" need to be adjusted, but the adjustment value of the "output power of the screwdriver" is larger, indicating that it is more critical for improving the tightening quality and the adjustment difficulty is moderate, then the "output power of the screwdriver" parameter is adjusted first. According to the determined adjustment priorities, sequentially adjust multiple target operating parameters of the screwdriver machine. During the adjustment process, follow the principle of starting from the easy ones and then moving to the difficult ones, and from the key ones to the secondary ones, to ensure that each adjustment step can effectively solve the main problems faced currently and gradually optimize the operating state of the screwdriver machine. After adjusting each operating parameter, monitor the changes in the data of the target sensors in real time, such as torque, displacement, temperature, etc. Through the real-time feedback of the sensor data, evaluate the effect of parameter adjustment to ensure that the adjusted parameters can achieve the expected improvement goals. For example, after adjusting the output power of the screwdriver, by monitoring the data of the torque sensor, observe whether the tightening torque has increased. If the increase amplitude meets the expectation, then continue to adjust the next parameter; if the increase is not obvious or other abnormal data changes occur, then it is necessary to re-evaluate and adjust the current parameter to ensure the accuracy and reliability of the adjustment process.

[0036] According to the relationship between the fault type and the operating parameters, accurately determine the target operating parameters corresponding to the diagnostic result. This precise positioning avoids ineffective adjustment of irrelevant parameters, improves the pertinence and efficiency of fault handling. By calculating the tightening quality influence coefficient and the adjustment difficulty coefficient of each target operating parameter, and performing weighted processing to obtain the adjustment value, and then determine the adjustment priority. Prioritize the adjustment of parameters that have a great impact on the tightening quality and relatively low adjustment difficulty. This priority allocation strategy makes the fault handling more orderly and efficient, can quickly solve key problems, and reduce the impact time of the fault on production. According to the weighted results of the tightening quality influence coefficient and the adjustment difficulty coefficient, calculate the precise adjustment value for each target operating parameter. This refined adjustment method can ensure that the change of each parameter has a positive impact on the tightening quality, avoids over-adjustment or under-adjustment that may be caused by extensive adjustment, thereby optimizing the tightening quality of the screws and improving the reliability and consistency of the product. During the process of adjusting the operating parameters, real-time monitor the changes in the data of the target sensors, such as torque, displacement, and temperature. If it is found that the adjusted parameters fail to achieve the expected effect or cause abnormalities in other parameters, dynamic adjustment can be immediately performed. This real-time monitoring and dynamic adjustment mechanism ensures the stability of the tightening process, ensures that the screwing machine is always in the best operating state, and further improves the tightening quality.

[0037] Optionally, the determining the preset production result according to the inventory data and the actual demand includes: Obtain the inventory quantity and inventory turnover rate of the target component, and predict the inventory demand for a future preset time period through the inventory quantity and the inventory turnover rate; Obtain the urgency and quantity requirements in the customer order information to determine the actual demand; Determine the preset production result through the inventory demand and the actual demand, and the preset production result includes the production beat, the production quantity target, and the product quality standard.

[0038] Obtain the inventory quantity of the target component, that is, the actual stock of components in the current warehouse; at the same time, collect the inventory turnover rate of the components, that is, the frequency of components entering and leaving the warehouse within a certain period. These two indicators are the core data of inventory management and can intuitively reflect the inventory status and flow of components. Based on the inventory quantity and inventory turnover rate, use forecasting algorithms (such as time series analysis, exponential smoothing method, etc.) to forecast the inventory demand of the target component within a preset time period in the future. For example, if the inventory turnover rate of a component is high, it indicates that its consumption speed is fast and more inventory may be needed in the future to meet production requirements; conversely, if the turnover rate is low, the future inventory demand is relatively small. The forecasting result provides a reference basis for the production plan in terms of inventory and helps to avoid the occurrence of inventory backlogs or shortages. The system obtains customer order information from the order management system, focusing on the urgency and quantity requirements of the orders. The urgency reflects the customer's urgent need for the product delivery time, such as an urgent order that needs to be produced first; the quantity requirement clarifies the customer's specific demand for the product quantity and is a key factor in determining the production scale. Based on the urgency and quantity requirements in the customer order information, comprehensively analyze and determine the actual demand. For example, if multiple customer orders are urgent and the quantity is large, it indicates that the market demand for the product is strong, and the production priority and output need to be increased; if the order quantity is small and not urgent, the production plan can be adjusted appropriately according to the actual situation to meet customer needs while reasonably arranging production resources. Combine the inventory demand and the actual demand to determine the production rhythm, that is, the number of production tasks completed per unit time. If the inventory demand is large and the customer demand is urgent, the production rhythm can be accelerated to improve production efficiency to meet market demand and order delivery requirements; if the inventory is sufficient and the customer demand is relatively stable, the normal production rhythm can be maintained to ensure the stability and continuity of production. According to the predicted value of the inventory demand and the quantity requirements of the customer orders, formulate a reasonable production volume target. The production volume target should not only meet the replenishment demand of the inventory but also meet the order demand of the customers, while considering the reasonable utilization of production capacity and resources to avoid overproduction or idle production capacity. When determining the preset production result, the product quality standard is crucial. According to the specific requirements of the customer order for product quality, combined with the enterprise's quality management system and industry standards, clarify the product quality standard. This includes all aspects such as the appearance quality, dimensional accuracy, and performance indicators of the product, ensuring that the products produced can meet customer expectations and market demands, and improving the market competitiveness of the products and the brand image of the enterprise.

[0039] By obtaining the inventory quantity and inventory turnover rate of target components, the inventory demand for a future preset time period can be accurately predicted. The inventory quantity reflects the current inventory level, while the inventory turnover rate reveals the consumption speed and frequency of inventory. Combining these two indicators, the required inventory quantity within a specific time can be reasonably predicted, providing a reliable inventory basis for production planning and avoiding production delays or resource waste caused by insufficient or excessive inventory. By obtaining the urgency level and quantity requirements in customer order information, the actual needs of customers can be directly understood. Orders with a high urgency level mean that customers have strict requirements for delivery time and need to be given priority in production; the quantity requirements specify the number of screws required by customers. By analyzing this order information, the dynamic market demand can be accurately grasped, ensuring that the production plan matches the market demand, and improving customer satisfaction and market response speed. According to the production rhythm in the preset production results, the resource allocation such as equipment operation, personnel scheduling, and raw material supply in the production process can be reasonably arranged. The determination of the production rhythm takes into account the balance between inventory demand and market demand, avoiding inventory backlogs caused by too fast production rhythm or order delays caused by too slow production rhythm. A reasonable production rhythm helps to improve equipment utilization rate, reduce production costs, and enhance production efficiency. The determination of the production output target is based on a comprehensive analysis of inventory demand and market demand. When the inventory is sufficient and the market demand is stable, the production output target can be appropriately reduced to reduce inventory costs and production pressure; while when the inventory is tight and the market demand is strong, the production output target is increased to meet the market demand and increase enterprise revenue. A scientific production output target helps to optimize the allocation of production resources and maximize production benefits. The determination of the preset production results takes into account various factors, making the production plan have a certain degree of flexibility. During the production process, in case of sudden situations such as equipment failures and raw material supply problems, the production rhythm, production output target, and product quality standards can be flexibly adjusted according to the actual situation to ensure the smooth progress of production and the timely delivery of orders, improving the flexibility and risk resistance of production.

[0040] Optionally, the comparing the production results with the preset production results to obtain a comparison result and adjusting the operating parameters according to the comparison result includes: Comparing the key indicators in the production results with the corresponding indicators in the preset production results one by one to determine the deviation type and deviation degree, where the key indicators include production output and product quality pass rate; Matching the optimal operating parameter adjustment plan from the preset parameter adjustment strategy library according to the deviation type and the deviation degree.

[0041] Select the production output and the qualified rate of product quality as key indicators for comparison. The production output directly reflects the production efficiency and the ability to meet market demand, while the qualified rate of product quality reflects the quality control level in the production process. These two indicators are the core factors for measuring whether the production process meets the expected goals. Compare the production output in the actual production results with the production output target in the preset production results to determine whether there is a deviation in the production output and the specific degree of the deviation, such as whether the production output is lower or higher than the target value and what the percentage of the deviation is. Similarly, compare the actual qualified rate of product quality with the preset product quality standard to judge whether the product quality meets the standard and the degree of the deviation, such as whether the qualified rate is lower than the preset standard and what the proportion of unqualified products is. Through the comparison and analysis, determine the type of deviation, such as whether the production output is insufficient or excessive, and whether the product quality has a low qualified rate or specific quality problems (such as loose screws, stripped threads, etc.). At the same time, quantify the degree of the deviation to provide an accurate basis for subsequent parameter adjustment. Establish a preset parameter adjustment strategy library containing multiple parameter adjustment schemes. Each scheme provides corresponding suggestions for adjusting the operating parameters for different types and degrees of deviation. For example, for the situation of insufficient production output, there may be adjustment schemes such as increasing the output power of the screwdriver and accelerating the tightening speed; for the situation of a low qualified rate of product quality, there may be schemes such as improving the positioning accuracy of the screwdriver machine and optimizing the screw feeding path. According to the determined type and degree of deviation, match the optimal operating parameter adjustment scheme from the preset parameter adjustment strategy library. The matching process takes into account factors such as the urgency of the deviation and the feasibility and effectiveness of the adjustment scheme to ensure that the selected scheme can most effectively solve the current production deviation problem. For example, if the production output is severely insufficient and the market demand is urgent, preferentially select an adjustment scheme that can quickly increase the production output, such as increasing the operating power and speed of multiple screwdriver machines at the same time. Adjust the operating parameters of the screwdriver machine according to the matched optimal adjustment scheme. During the adjustment process, monitor the changes in the target sensor data in real time to ensure that the parameter adjustment achieves the expected effect. If the production results still do not meet the preset goals after the adjustment, the system will conduct a comparison analysis and parameter adjustment again until the production process is stable and the production results meet the preset requirements.

[0042] By comparing key indicators in production results (such as output and product quality pass rate) with corresponding indicators in pre-set production results, the type and extent of deviations occurring during the production process can be accurately identified. This precise identification helps pinpoint production process issues and provides clear direction and basis for subsequent parameter adjustments. Based on the identified deviation type and extent, the optimal operating parameter adjustment plan is selected from a library of pre-set parameter adjustment strategies. This timely adjustment mechanism rapidly responds to changes in the production process, prevents deviations from escalating, ensures the production process remains under control, and improves production stability and reliability. The comparison of production results with pre-set results and the selection of adjustment plans from the pre-set parameter adjustment strategy library are based on extensive production data. This data-driven decision support approach makes production management more scientific and precise, reduces the subjectivity and uncertainty of human decision-making, and enhances the intelligence of production management. By continuously comparing production results with pre-set targets and adjusting operating parameters based on these comparisons, a continuous optimization cycle is formed. This continuous optimization mechanism enables the production process to continuously improve efficiency and quality, thereby providing companies with sustained efficiency gains and competitive advantages.

[0043] To ensure the system's high performance and low latency, all hardware utilizes industrial-grade quality products. For example, the high-definition cameras utilize Hikvision's DS-2CD2646G0-I8 series IP cameras, and the data processing components are powered by Intel Core i7-class CPU chipsets. Software leverages open-source frameworks such as TensorFlow to train deep learning models, optimizing object positioning accuracy to over 95%. A MySQL database is also used to store long-term historical records for easy auditing and future queries. If an unexpected situation occurs, such as motor overheating or bearing wear, the emergency brake activates, halting all activity until the situation is resolved. Finally, if a danger signal is not properly addressed, the ultimate protective barrier activates, continuously sounding the horn and flashing a red light until the crisis is resolved.

[0044] This embodiment also discloses a screw machine control and management system. Figure 2 This is a module diagram of the screw machine control management system disclosed in the embodiment of this application, such as Figure 2 As shown, the system includes a collection module 201, an integration module 202, an adjustment module 203 and an execution module 204, wherein: An acquisition module 201 is configured to acquire target images, target sensor data, and target sounds, wherein the target images include the installation position and posture of the screw and images of the screw during tightening; the target sensor data include torque, displacement, and temperature data during the screw tightening process; and the target sounds include sounds generated by the screw machine during operation. an integration module 202 configured to integrate the target image, the target sensor data, and the target sound to obtain target data, and process the target data using a preset deep learning model to obtain a diagnosis result; An adjustment module 203 is configured to adjust operating parameters of the screw machine according to the diagnosis result and the target sensor data, and obtain a production result of the screw machine after adjustment, wherein the operating parameters include the output power of the screwdriver, the tightening speed, and the positioning accuracy of the screw machine; The execution module 204 is configured to determine a preset production result according to inventory data and actual demand, compare the production result with the preset production result to obtain a comparison result, and adjust the operating parameters according to the comparison result.

[0045] Optionally, the integration module 202 is configured to: preprocessing the target image, the target sensor data, and the target sound, and synchronizing the target image, the target sensor data, and the target sound in time by interpolation; extracting visual features from the preprocessed target image, the visual features including edge features and texture features, extracting physical features from the preprocessed target sensor data, the physical features including moment mean and displacement variance, and extracting audio features from the preprocessed target sound, the audio features including spectral features and time domain features; The visual features, the physical features, and the audio features are fused to construct a comprehensive feature vector, and the comprehensive feature vector is used as the target data.

[0046] Optionally, the integration module 202 is configured to: Combining visual features with physical features, the first correlation between the appearance quality of screws and the tightening effect is analyzed through the connection of neurons and the nonlinear transformation of activation functions; Combining visual features with audio features to analyze the secondary correlation between the appearance of the screw and the operating sound; Combining physical features with audio features, and analyzing the third relationship between the physical changes and sound performance during the screw tightening process through hidden layer calculations; A diagnosis result is determined according to the first correlation, the second correlation, and the third correlation.

[0047] Optionally, the integration module 202 is configured to: performing weighted processing on the first association, the second association, and the third association according to a preset weight distribution mechanism to obtain a comprehensive diagnostic score; If the comprehensive diagnosis score is greater than or equal to a preset threshold, output a diagnosis result indicating that the screw tightening is normal; If the comprehensive diagnosis score is less than the preset threshold, output a fault type according to the comprehensive diagnosis score.

[0048] Optionally, the adjustment module 203 is configured to: Determine target operating parameters corresponding to the diagnosis result according to the relationship between the fault type and the operating parameters, and determine the tightening quality influence coefficient and the adjustment difficulty coefficient for each of the target operating parameters; Perform weighted processing according to the tightening quality influence coefficient and the adjustment difficulty coefficient to obtain an adjustment value for each of the target operating parameters, and determine the adjustment priority for the multiple target operating parameters according to the adjustment value; Adjust the multiple target operating parameters according to the adjustment priority, and monitor the change of the target sensor data in real time.

[0049] Optionally, the execution module 204 is configured to: Obtain the inventory quantity and inventory turnover rate of the target component, and predict the inventory demand for a future preset time period through the inventory quantity and the inventory turnover rate; Obtain the urgency and quantity requirements in the customer order information to determine the actual demand; Determine a preset production result through the inventory demand and the actual demand, and the preset production result includes a production beat, a production quantity target, and a product quality standard.

[0050] Optionally, the execution module 204 is configured to: Compare the key indicators in the production result with the corresponding indicators in the preset production result one by one to determine the deviation type and the deviation degree, and the key indicators include the production quantity and the product quality qualification rate; Match the optimal operating parameter adjustment plan from the preset parameter adjustment strategy library according to the deviation type and the deviation degree.

[0051] It should be noted that: when the device provided in the above embodiment realizes its functions, only the above-mentioned division of each functional module is used for illustration. In actual application, the above functions can be allocated to different functional modules according to needs, that is, the internal structure of the device is divided into different functional modules to complete all or part of the functions described above. In addition, the device and method embodiments provided in the above embodiments belong to the same concept, and the specific implementation process can be seen in the method embodiment, which will not be elaborated here.

[0052] This embodiment also discloses an electronic device, referring to Figure 3, the electronic device may include: at least one processor 301, at least one communication bus 302, a user interface 303, a network interface 304, and at least one memory 305.

[0053] Among them, the communication bus 302 is used to realize the connection and communication between these components.

[0054] Among them, the user interface 303 may include a display screen (Display) and a camera (Camera). Optionally, the user interface 303 may further include a standard wired interface and a wireless interface.

[0055] Among them, the network interface 304 may optionally include a standard wired interface and a wireless interface (such as a WI-FI interface).

[0056] Among them, the processor 301 may include one or more processing cores. The processor 301 connects various parts within the entire server through various interfaces and lines. By running or executing instructions, programs, code sets, or instruction sets stored in the memory 305, and by calling the data stored in the memory 305, it performs various functions of the server and processes data. Optionally, the processor 301 may be implemented in at least one hardware form of digital signal processing (DSP), field-programmable gate array (FPGA), or programmable logic array (PLA). The processor 301 may integrate one or several combinations of a central processing unit (CPU), a graphics processing unit (GPU), and a modem, etc. Among them, the CPU mainly processes the operating system, the user interface, and application programs, etc.; the GPU is responsible for the rendering and drawing of the content to be displayed on the display screen; the modem is used to process wireless communication. It can be understood that the above modem may not be integrated into the processor 301 and may be implemented separately through a single chip.

[0057] Among them, the memory 305 may include a Random Access Memory (RAM), or may also include a Read-Only Memory. Optionally, the memory 305 includes a non-transitory computer-readable storage medium. The memory 305 can be used to store instructions, programs, codes, code sets or instruction sets. The memory 305 may include a program storage area and a data storage area. Among them, the program storage area may store instructions for implementing the operating system, instructions for at least one function (such as touch function, sound playback function, image playback function, etc.), instructions for implementing the above-mentioned method embodiments, etc.; the data storage area may store the data involved in the above-mentioned method embodiments. Optionally, the memory 305 may also be at least one storage device located far from the aforementioned processor 301. As Figure 3 shown, the memory 305, as a computer storage medium, may include an operating system, a network communication module, a user interface module, and an application program for the screw machine control and management method.

[0058] In Figure 3 the electronic device shown, the user interface 303 is mainly used to provide an input interface for the user to obtain the data input by the user; and the processor 301 can be used to call the application program for the screw machine control and management method stored in the memory 305. When executed by one or more processors 301, the electronic device is caused to execute one or more of the methods in the above embodiments.

[0059] It should be noted that for the foregoing method embodiments, for the sake of simple description, they are all expressed as a series of action combinations. However, those skilled in the art should know that this application is not limited by the described action sequence, because according to this application, certain steps can be performed in other sequences or simultaneously. Secondly, those skilled in the art should also know that the embodiments described in the specification are all preferred embodiments, and the actions and modules involved are not necessarily required by this application.

[0060] In the above embodiments, the descriptions of the respective embodiments have their own emphases. For the parts not detailed in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0061] In several embodiments provided in this application, it should be understood that the disclosed device can be implemented in other ways. For example, the device embodiments described above are merely illustrative. For example, the division of units is only a logical function division. In actual implementation, there may be other division methods. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the displayed or discussed coupling or direct coupling or communication connection between each other can be through some service interfaces. The indirect coupling or communication connection of the device or unit can be in electrical or other forms.

[0062] The units described as separate components may or may not be physically separated. The components displayed as units may or may not be physical units, that is, they can be located in one place or distributed to multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0063] In addition, each functional unit in various embodiments of this application can be integrated in a processing unit, or each unit can exist physically alone, or two or more units can be integrated in one unit. The above-mentioned integrated units can be implemented in the form of hardware or in the form of software functional units.

[0064] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable memory. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of this technical solution, can be embodied in the form of a software product. This computer software product is stored in a memory 305 and includes several instructions to enable a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods in various embodiments of this application. And the aforementioned memory 305 includes: various media such as USB flash drives, mobile hard disks, magnetic disks, or optical discs that can store program codes.

[0065] The above are only exemplary embodiments of the present disclosure and cannot be used to limit the scope of the present disclosure. That is, all equivalent changes and modifications made in accordance with the teachings of the present disclosure still fall within the scope covered by the present disclosure. Those skilled in the art will easily think of other implementation schemes of the present disclosure after considering the disclosure of the specification. This application aims to cover any variations, uses, or adaptive changes of the present disclosure. These variations, uses, or adaptive changes follow the general principles of the present disclosure and include common general knowledge or conventional technical means in the technical field not recorded in the present disclosure. The specification and the embodiments are only regarded as exemplary, and the scope and spirit of the present disclosure are defined by the claims.

Claims

1. A control and management method for a screwdriver machine, characterized in that, Applied to a screwdriver machine, the method includes: Obtain a target image, target sensor data, and target sound. The target image includes the installation position and posture of the screw and images during the screw tightening process. The target sensor data includes torque, displacement, and temperature data during the screw tightening process. The target sound includes the sound generated by the screwdriver machine during operation; Integrate the target image, the target sensor data, and the target sound to obtain target data, and process the target data through a preset deep learning model to obtain a diagnosis result; Adjust the operating parameters of the screwdriver machine according to the diagnosis result and the target sensor data, and obtain the production result of the adjusted screwdriver machine. The operating parameters include the output power of the screwdriver, the tightening speed, and the positioning accuracy of the screwdriver machine; Determine a preset production result according to inventory data and actual demand, compare the production result with the preset production result to obtain a comparison result, and adjust the operating parameters according to the comparison result.

2. The screw machine control and management method according to claim 1, wherein The integrating the target image, the target sensor data, and the target sound to obtain target data includes: Preprocess the target image, the target sensor data, and the target sound, and synchronize the target image, the target sensor data, and the target sound in time through interpolation; Extract visual features from the preprocessed target image. The visual features include edge features and texture features. Extract physical features from the preprocessed target sensor data. The physical features include torque mean and displacement variance. Extract audio features from the preprocessed target sound. The audio features include spectral features and time-domain features; Fuse the visual features, the physical features, and the audio features to construct a comprehensive feature vector, and use the comprehensive feature vector as the target data.

3. The screw machine control and management method according to claim 1, wherein, The processing the target data through a preset deep learning model to obtain a diagnosis result includes: Combine the visual features with the physical features, and analyze the first correlation between the appearance quality of the screw and the tightening effect through the connection of neurons and the nonlinear transformation of the activation function; Combine the visual features with the audio features, and analyze the second correlation between the appearance state of the screw and the operating sound; Combine the physical features with the audio features, and analyze the third correlation between the physical changes and the sound performance during the screw tightening process through the calculation of the hidden layer; Determine the diagnosis result according to the first correlation, the second correlation, and the third correlation.

4. The screw machine control and management method according to claim 3, characterized in that The determining the diagnosis result according to the first correlation, the second correlation, and the third correlation includes: Perform weighted processing on the first correlation, the second correlation, and the third correlation according to a preset weight distribution mechanism to obtain a comprehensive diagnosis score; If the comprehensive diagnosis score is greater than or equal to a preset threshold, output a diagnosis result indicating that the screw tightening is normal; If the comprehensive diagnosis score is less than the preset threshold, output the type of fault according to the comprehensive diagnosis score.

5. The method for controlling and managing a screwdriver machine according to claim 1, characterized in that, The adjusting the operating parameters of the screwdriver machine according to the diagnosis result and the target sensor data includes: Determine the target operating parameters corresponding to the diagnosis result according to the relationship between the fault type and the operating parameters, and determine the tightening quality influence coefficient and the adjustment difficulty coefficient of each of the target operating parameters; Perform weighted processing according to the tightening quality influence coefficient and the adjustment difficulty coefficient to obtain the adjustment value of each of the target operating parameters, and determine the adjustment priority of the multiple target operating parameters according to the adjustment value; Adjust the multiple target operating parameters according to the adjustment priority, and monitor the change of the target sensor data in real time.

6. The screw machine control and management method according to claim 1, characterized in that The determining the preset production result according to the inventory data and the actual demand includes: Obtain the inventory quantity and inventory turnover rate of the target component, and predict the inventory demand in a future preset time period through the inventory quantity and the inventory turnover rate; Obtain the urgency and quantity requirements in the customer order information to determine the actual demand; Determine the preset production result through the inventory demand and the actual demand, and the preset production result includes the production rhythm, the output target and the product quality standard.

7. The screw machine control and management method according to claim 6, characterized in that, The comparing the production result with the preset production result to obtain a comparison result, and adjusting the operating parameters according to the comparison result includes: Compare the key indicators in the production result with the corresponding indicators in the preset production result one by one to determine the deviation type and the deviation degree, and the key indicators include the output and the product quality pass rate; Match the optimal operating parameter adjustment plan from the preset parameter adjustment strategy library according to the deviation type and the deviation degree.

8. A screwdriver control and management system, characterized in that, It includes a collection module, an integration module, an adjustment module and an execution module, wherein: The collection module is configured to obtain a target image, target sensor data and target sound, the target image includes the installation position and posture of the screw and the image during the screw tightening process, the target sensor data includes the torque, displacement and temperature data during the screw tightening process, and the target sound includes the sound generated by the screw machine during operation; The integration module is configured to integrate the target image, the target sensor data and the target sound to obtain target data, and process the target data through a preset deep learning model to obtain a diagnosis result; The adjustment module is configured to adjust the operating parameters of the screw machine according to the diagnosis result and the target sensor data, and obtain the production result of the adjusted screw machine, and the operating parameters include the output power of the screwdriver, the tightening speed and the positioning accuracy of the screw machine; The execution module is configured to determine the preset production result according to the inventory data and the actual demand, compare the production result with the preset production result to obtain a comparison result, and adjust the operating parameters according to the comparison result.

9. An electronic device, characterized in that, It includes a processor, a memory, a user interface and a network interface, the memory is used to store instructions, both the user interface and the network interface are used to communicate with other devices, and the processor is used to execute the instructions stored in the memory so that the electronic device executes the method according to any one of claims 1-7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores instructions that, when executed, perform the method according to any one of claims 1-7.

Images