Screw locking state determination method and device, electronic equipment and storage medium

Abstract

The invention discloses a method and device for determining a screw locking state, electronic equipment and a storage medium, belongs to the technical field of workpiece locking, and is used for improving the accuracy of screw locking state detection. The method comprises the following steps: acquiring first working data of a screw locking working end in a first screw locking process; the first working data are input into a target model, the screw locking state is output through the target model, and the target model comprises the corresponding relation between the first working data and the screw locking state.

Description

Technical Field

[0001] The present application belongs to the technical field of workpiece locking, and specifically relates to a method, device, electronic device and storage medium for determining the locking state of a screw. Background Art

[0002] With the development and advancement of science and technology, the "screw locking" process in the communications, automotive, and mechanical assembly industries is gradually being replaced by automated equipment. While this process requires high precision and strong repeatability, manual screw locking is susceptible to numerous external factors during production, leading to problems such as screw misalignment, thread slippage, and re-entry, impacting the safety and lifespan of the manufactured product. Against this backdrop, automated screw locking is rapidly evolving, continuously developing, and improving, bringing with it the need for monitoring and early warning systems for this process.

[0003] Currently, the method for detecting screw locking mainly relies on manual visual inspection in the post-process, but manual inspection is inefficient and still has problems such as missed detection, false detection, and inability to accurately detect the screw locking status. Summary of the Invention

[0004] The embodiments of the present application provide a method, device, electronic device, and storage medium for determining the locking state of a screw, which can solve the problem of being unable to accurately detect the locking state of a screw.

[0005] In a first aspect, an embodiment of the present application provides a method for determining a screw locking state, the method comprising: obtaining first working data of a screw locking working end during a first screw locking process; inputting the first working data into a target model, and outputting the screw locking state through the target model, wherein the target model includes a correspondence between the first working data and the screw locking state.

[0006] In a second aspect, an embodiment of the present application provides a device for determining a screw locking state, the device comprising: an acquisition module for acquiring first working data of a screw locking working end during a first screw locking process; a determination module for inputting the first working data into a target model and outputting the screw locking state through the target model, the target model including a correspondence between the first working data and the screw locking state.

[0007] In a third aspect, an embodiment of the present application provides an electronic device comprising a processor, a memory, and a program or instruction stored in the memory and executable on the processor, wherein the program or instruction, when executed by the processor, implements the steps of the method described in the first aspect.

[0008] In a fourth aspect, an embodiment of the present application provides a readable storage medium, on which a program or instruction is stored. When the program or instruction is executed by a processor, the steps of the method described in the first aspect are implemented.

[0009] In an embodiment of the present application, by obtaining the first working data of the screw locking working end during the first screw locking process; inputting the first working data into a target model, and outputting the screw locking state through the target model, the target model includes a correspondence between the first working data and the screw locking state, thereby realizing automatic determination of the screw locking state, avoiding the low efficiency of manual detection and the problem of missed detection, wrong detection, and the inability to accurately detect the screw locking state, thereby improving the accuracy of the screw locking state detection. BRIEF DESCRIPTION OF THE DRAWINGS

[0010] Figure 1 1 is a flow chart of a method for determining a screw locking state provided in an embodiment of the present application;

[0011] Figure 2 This is an interpolation intention provided by the embodiment of the present application;

[0012] Figure 3 This is a schematic diagram of the rotation angle-torque timing of a screw locking state provided by an embodiment of the present application;

[0013] Figure 4 1 is a flow chart of another method for determining a screw locking state provided in an embodiment of the present application;

[0014] Figure 5 1 is a schematic structural diagram of a device for determining a screw locking state provided in an embodiment of the present application;

[0015] Figure 6 This is a schematic diagram of the structure of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0016] The following will be combined with the drawings in the embodiments of this application to clearly and completely describe the technical solutions in the embodiments of this application. Obviously, the embodiments described are part of the embodiments of this application, not all of them. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of this application.

[0017] The terms "first," "second," and the like in the specification and claims of this application are used to distinguish similar objects, and are not used to describe a specific order or precedence. It should be understood that the terms used in this manner are interchangeable where appropriate, so that the embodiments of this application can be implemented in an order other than that illustrated or described herein, and that the objects distinguished by "first," "second," and the like are generally of the same type, and do not limit the number of objects; for example, the first object can be one or more. In addition, the term "and / or" in the specification and claims refers to at least one of the connected objects, and the character " / " generally indicates that the objects connected are in an "or" relationship.

[0018] The following describes in detail the method, device, electronic device, and storage medium for determining the screw locking state provided by the embodiments of the present application through specific embodiments and their application scenarios in conjunction with the accompanying drawings.

[0019] Figure 1 A method for determining the screw locking state provided by an embodiment of the present application is shown. The method can be executed by an electronic device. In other words, the method can be executed by software or hardware installed in the electronic device. The method includes the following steps:

[0020] Step 102: Acquire first working data of the screw locking working end during the first screw locking process.

[0021] In this embodiment of the present application, fault diagnosis and self-repair of the screw locking status primarily include steps such as data acquisition, data preprocessing, model design and training, fault classification, and self-repair. The screw locking working end primarily consists of a travel mechanism, a screw locking module, and a sensor. The screw locking module is mounted at the end of the travel mechanism to achieve screw locking. During the first screw locking process, data acquisition primarily involves sensors capturing the first operational data of the screw locking working end during the first screw locking process. Data acquisition is performed by sensors using industrial communication protocols to collect and transmit signals.

[0022] In one implementation, the first working data includes rotation angle data and torque data of the screw locking working end during the first screw locking process.

[0023] Specifically, the sensor collects the screw locking module's rotation angle and torque data during the first locking process, namely, sequential rotation angle and sequential torque data. The sensor and reverse data are collected for the screw locking working end. This data is transmitted using industrial communication protocols such as Modbus, RS-232, and HART.

[0024] Step 104: inputting the first working data into a target model, and outputting a screw locking state through the target model, wherein the target model includes a correspondence between the first working data and the screw locking state.

[0025] Specifically, after collecting the first working data, the first working data can be input into a target model, and the target model can be used to diagnose and output the screw locking state of the first screw locking process. The target model is pre-trained and includes a correspondence between the first working data and the screw locking state.

[0026] The method for determining the screw locking state provided in an embodiment of the present application obtains first working data of a screw locking working end during a first screw locking process; inputs the first working data into a target model, and outputs the screw locking state through the target model. The target model includes a correspondence between the first working data and the screw locking state, thereby enabling automatic determination of the screw locking state, avoiding the low efficiency of manual detection and the problem of missed detection, false detection, and the inability to accurately detect the screw locking state, thereby improving the accuracy of screw locking state detection.

[0027] In one implementation, before obtaining the first working data of the screw locking working end in the first screw locking process, the method further includes:

[0028] During the process of training multiple models, a first quantity of training data corresponding to the correct screw locking state output by each model is obtained; a first ratio is determined based on the first quantity and the second quantity of all training data; during the process of testing multiple models, a third quantity of test data corresponding to the correct screw locking state output by each model is obtained; a second ratio is determined based on the third quantity and the fourth quantity of all test data; and the target model is determined from the multiple models based on the first ratio and the second ratio.

[0029] In an embodiment of the present application, multiple models are provided for diagnosing screw locking status. For example, the multiple models can be a Multilayer Perceptron (MLP) model, a Convolutional Neural Network (CNN) model, or a Long Short-Term Memory (LSTM) model. In this embodiment of the present application, the performance of the models is compared using accuracy. Based on the ratio of the number of correctly classified fault samples in the training model to the total number of fault samples, a fault state classification accuracy (FS Accuracy) is designed.

[0030] Specifically, during the training of multiple models, the optimal fault diagnosis model for a specific locking scenario is selected and confirmed by using the ratio of the number of correctly classified samples in the training data used to train the models to the total number of training data. Specifically, a first number of training data corresponding to the correct screw locking state output by each model is obtained, and a first ratio between the first number and a second number of all training data is determined. During the testing of multiple models, a third number of test data corresponding to the correct screw locking state output by each model is obtained, and a second ratio between the third number and a fourth number of all test data is determined. Based on the first and second ratios, a target model is determined from the multiple models. Typically, the model with the highest first and second ratios is determined as the optimal fault diagnosis model, i.e., the target model described above. In this way, the designed models are trained using the training data, and the optimal target model for a specific locking scenario is selected and confirmed by using the ratio of the number of correctly classified samples in the training data to the total number of samples in the training data. The target model can then output the screw locking state, enabling automated determination of the screw locking state. This avoids the inefficient manual detection process, which can still lead to missed or false detections, and other issues that can lead to inaccurate screw locking state detection, thereby improving the accuracy of screw locking state detection.

[0031] In one implementation, obtaining first working data of the screw locking working end during the first screw locking process includes:

[0032] Acquire second working data of the screw locking working end during the first screw locking process; uniformly interpolate the second working data into multiple interpolation data along the acquisition time of the second working data; uniformly sample the interpolated second working data; and normalize the second working data obtained by uniform sampling to obtain the first working data.

[0033] Specifically, to obtain first operating data of the screw locking working end during the first screw locking process, it is first necessary to obtain second operating data of the screw locking working end during the first screw locking process. The second operating data is preprocessed to obtain the first operating data. The second operating data may include time-series rotation angle data and time-series torque data during the first screw locking process.

[0034] The second operating data can be obtained by filtering out noise signals through isochronous piecewise linear interpolation, uniform sampling, and normalization to obtain the first operating data. Specifically, the second operating data can be represented by A2n. The first dimension of the second operating data consists of two features: time-series angle and time-series torque. The second dimension n is the time-series length. That is, the locking data consists of several (Angle, Torque) two-dimensional point pairs. Furthermore, the operating data of the locking process features isochronous sampling and increased angle variation.

[0035] The initial sampling time of the second working data can be regarded as [1,n]. K points are uniformly interpolated along this time dimension, and the angle interpolation points and torque interpolation points are obtained by piecewise linear interpolation. The mathematical model is as follows:

[0036]

[0037] Among them, t in the above mathematical model i ’ represents the time of the i-th interpolation point, x i ’ Indicates the rotation angle of the i-th interpolation point, y i ’ represents the torque at the i-th interpolation point, Figure 2 An interpolation schematic diagram provided by an embodiment of the present application is shown. Uniform sampling is performed from the interpolated second working data, and each second working data is converted into a sequence of equal length data. The maximum torque and rotation angle of the screw locking in different positions and under different conditions of the equal length sequence data are different. The second working data obtained by uniform sampling are normalized to obtain the first working data. The mathematical model of the normalization process is as follows:

[0038]

[0039]

[0040] Among them, Angle represents the angle data, Torque represents the torque data, and NormalPeakTorque represents the normal peak torque.

[0041] In this way, the second working data is subjected to the isochronous piecewise linear interpolation, uniform sampling and normalization method, which can filter out and shield the mixed signals to obtain the first working data, thereby improving the accuracy of the locking state diagnosis.

[0042] In one implementation, after outputting the screw locking state through the target model, the method further includes:

[0043] When the screw locking state output by the target model is a fault state, the fault is repaired according to the correction strategy corresponding to the fault state; wherein, when the fault state is slipped thread locking, the correction strategy is to adjust the locking angle of the screw locking working end, and the fault is repaired according to the torque-angle state of the screw locking working end; when the fault state is repeated locking, the correction strategy is to absorb the repeated screws and remove the repeatedly locked screws; when the fault state is oblique locking, the correction strategy is to control the direction, locking angle and torque of the screw locking working end, reversely loosen the screw and adjust the locking angle.

[0044] Specifically, the second working data of screw locking collected by the sensor in real time is pre-processed and detected by the optimal fault diagnosis model to determine whether it is normal, slipped, tilted, repeated, etc.

[0045] According to the screw locking state obtained by fault diagnosis, the corresponding correction strategy can be adaptively matched and selected to repair the fault. Specifically, in the case where the fault state is slipped thread locking, the correction strategy is to adjust the locking angle of the screw locking working end, readjust the locking angular velocity to engage the screw, and perform adaptive fault repair according to the torque-angle state of the screw locking working end; in the case where the fault state is repeated locking, the correction strategy is to suck up the repeated screws through the air nozzle / electromagnetic suction and remove the repeatedly locked screws; in the case where the fault state is oblique locking, the correction strategy is to control the direction, locking angle and torque of the screw locking working end, reversely loosen the screw and adjust the locking angle to achieve adaptive repair of the oblique state type of locking fault. In this way, different screw repair strategies are executed according to different screw locking states, which can achieve the function of self-repair.

[0046] In one implementation, before obtaining the first working data of the screw locking working end in the first screw locking process, the method further includes:

[0047] Obtain third working data of the screw locking working end during the second screw locking process; perform a clustering operation on the third working data to obtain a cluster data set; analyze the cluster data set, and determine the correspondence between the screw locking state and the first working data based on the analysis results. The correspondence between the screw locking state and the first working data is used to train the target model, and the cluster data set includes the first working data.

[0048] In this embodiment of the present application, it is necessary to determine the correspondence between the screw locking state and the first operating data. In this embodiment of the present application, the third operating data of the second screw locking process, such as the time-series rotation angle and time-series torque, obtained through data acquisition, which can also be called historical data, is subjected to a K-means algorithm cluster analysis. The silhouette coefficient (SC) is used to evaluate the clustering results. Combined with the characteristics of the screw locking process, the fault type is classified into the early stage state.

[0049] Specifically, third working data of the screw locking working end in the second screw locking process is obtained, and a clustering operation is performed on the third working data to obtain a cluster data set.

[0050] In one implementation, performing a clustering operation on the third working data to obtain a clustered data set includes:

[0051] The third working data is divided into multiple clusters; the distance between the third working data and the class center of each cluster is repeatedly calculated and the third data is divided into the cluster corresponding to the class center closest to the third working data until the sum of the squares of the distances between each third working data and each class center does not change, thereby obtaining multiple cluster data sets.

[0052] Specifically, the third working data is divided into multiple clusters, for example, K clusters. The center of each cluster is calculated, as well as the sum of squared distances Je of all third working data to the cluster center. The distance to the cluster center is calculated for each third working data point, and the third working data point is reclassified into the cluster with the closest distance. This step is repeated until the sum of squared distances Je of each third working data point to each cluster center remains unchanged, resulting in multiple cluster data sets.

[0053] After obtaining a plurality of cluster data sets, the cluster data sets need to be analyzed. According to the analysis results, the corresponding relationship between the screw locking state and the first working data is determined, and the third working data includes the first working data.

[0054] In one implementation, analyzing the cluster data set includes:

[0055] The cluster data sets are analyzed according to the total number of the third working data, the average distance between the third working data in each cluster data set, and the average distance between each third working data and the third working data in the closest cluster data set.

[0056] In an embodiment of the present application, a mathematical model of the silhouette coefficient is provided as follows:

[0057]

[0058] In the silhouette coefficient mathematical model, N represents the total number of third training data points, ai represents the average distance between the i-th third training data point and other third training data points of the same type, and bi represents the average distance between the i-th third training data point and the third training data points in the closest cluster. A larger silhouette coefficient value indicates a smaller intra-class distance, while a larger inter-class distance indicates a clearer classification result. Thus, based on the results of the silhouette coefficient mathematical model, an appropriate number of fault classifications and states can be selected. Screw lock fault states can be generally categorized into normal, slipped, tilted, and repeated states. Figure 3 The following is a schematic diagram of the angle-torque timing of a screw locking state provided by an embodiment of the present application, wherein (a) represents the normal state, (b) represents the oblique locking position, (c) represents the slipping thread, and (d) represents the repeated locking. Figure 3 The corresponding relationship between the working data of the screw locking and the screw locking status can be seen in the figure.

[0059] The following is a detailed description of a method for determining a screw locking state provided by the present application through a specific embodiment. Figure 4 As shown, the method for determining the screw locking state includes the following process:

[0060] The method for determining the screw locking state includes screw locking, data collection, data preprocessing, model design and training, fault classification and self-repair. The screw locking working end is mainly composed of a stroke mechanism, a screw locking module and a sensor to achieve the screw locking work. The data collection is carried out by the sensor through the industrial communication protocol to collect and transmit the signal. The data is collected by A 2n The collected data are preprocessed by an optimized algorithm. The preprocessed data are converted into the probability distribution of state classes through a hidden layer composed of linear mapping and nonlinear activation function in the designed and trained model, and then the screw locking fault state is diagnosed and classified. Self-repair mainly selects the corresponding self-repair strategy based on the diagnosis and classification of different locking fault states and negatively feedback controls the screw locking tool end for correction.

[0061] Among them, the screw locking working end consists of a stroke mechanism, a screw locking module and a sensor. The screw locking module is installed at the end of the stroke mechanism. The sensor collects data such as the timing angle, timing torque, etc. of the screw locking module during the locking process.

[0062] The data such as the locking timing angle and timing torque of the screw locking working end are collected through sensors and retrograde, and the data collection adopts industrial communication protocols such as Modbus, RS-232, and HART for data transmission.

[0063] The historical data such as the time-series rotation angle and time-series torque obtained from data collection were clustered and analyzed using the K-means algorithm. The silhouette coefficient (SC) was used to evaluate the clustering results. Combined with the characteristics of the screw locking process, the fault types were divided into early stages.

[0064] In data collection, the sample set is divided into k clusters, and the centers of each class and the sum of the squares of the distances from all samples to the class center are calculated. e .

[0065] The distance between each sample and the class center is calculated, and the samples are reclassified into the class with the closest distance.

[0066] Repeat the above two steps until J e No change occurs.

[0067] The mathematical model of the silhouette coefficient of the sample set is as follows:

[0068]

[0069] In the silhouette coefficient mathematical model, N represents the total number of samples, a i represents the average distance between the i-th sample and other samples of the same type, b i It represents the average distance between the i-th sample and the samples in the closest cluster. A larger value of the silhouette coefficient indicates a smaller intra-class distance and a larger inter-class distance, which leads to a clearer classification result.

[0070] According to the calculation results of the contour coefficient mathematical model, the appropriate number and state of fault divisions are selected. The fault state can be basically divided into normal, slipped tooth, oblique position, repetitive state, etc.

[0071] Data acquisition is preprocessed by isochronous piecewise linear interpolation, uniform sampling and normalization methods to filter out and shield mixed signals.

[0072] Data collection samples are available 2n It means that the first dimension of the data sample is composed of two features: time series angle and time series torque. The second dimension n is the time series length, that is, the locked data sample is composed of several (Angle, Torque) two-dimensional point pairs. At the same time, the data has the characteristics of isochronous sampling and increasing angle speed.

[0073] The initial sampling time can be regarded as [1,n]. K points are uniformly interpolated along this time dimension. The angle interpolation points and torque interpolation points are obtained by piecewise linear interpolation. The mathematical model is as follows:

[0074]

[0075] In the mathematical model, t i ’ represents the time of the i-th interpolation point, x i ’ Indicates the rotation angle of the i-th interpolation point, y i ’ represents the torque at the i-th interpolation point.

[0076] The data is uniformly sampled, converting each sample into sequence data of equal length.

[0077] The maximum torque and angle of screw locking at different points and under different conditions of equal-length sequence data are different. After normalization, the mathematical model is as follows:

[0078]

[0079]

[0080] Angle means angle, Torque means torque, and NormalPeakTorque means normal peak torque.

[0081] Based on the preprocessed data, the multi-layer perceptron model MLP, convolutional neural network model CN ​​and long short-term memory model LSTM are designed to perform data processing and comparison to select the optimal fault diagnosis model strategy for specific locking scenarios.

[0082] The preprocessed data is trained on the designed model strategy, and the ratio of the number of correctly classified samples in the fault samples to the total number of fault samples is used to select and confirm the optimal fault diagnosis model for the specific locking scenario.

[0083] The screw locking data collected by the sensor in real time is pre-processed and tested by the optimal fault diagnosis model to determine whether it is normal, slipped, tilted, repeated, etc.

[0084] The screw locks the working end, and negative feedback is performed to control the inclination angle, torque, and speed of the locking working end according to different fault states to perform self-repair, and the above steps are repeated until the locking is normal.

[0085] It should be noted that the method for determining the screw locking state provided in the embodiments of the present application can be executed by a device for determining the screw locking state, or a control module within the device for determining the screw locking state that is used to execute the method. In the embodiments of the present application, the method for determining the screw locking state is executed by the device for determining the screw locking state as an example to illustrate the device for determining the screw locking state provided in the embodiments of the present application.

[0086] Figure 5 FIG. 1 is a schematic diagram of a device for determining a screw locking state according to an embodiment of the present application. Figure 5 As shown, the device 500 for determining the screw locking state includes: an acquisition module 510 and a determination module 520 .

[0087] An acquisition module 510 is used to obtain first working data of the screw locking working end during the first screw locking process; a determination module 520 is used to input the first working data into a target model and output the screw locking state through the target model, wherein the target model includes a correspondence between the first working data and the screw locking state.

[0088] In one implementation, the acquisition module 510 is further used to obtain, during the process of training multiple models, a first quantity of training data corresponding to the correct screw locking state output by each of the models; determine a first ratio based on the first quantity and a second quantity of all the training data; during the process of testing multiple models, obtain a third quantity of test data corresponding to the correct screw locking state output by each of the models; determine a second ratio based on the third quantity and a fourth quantity of all the test data; and determine the target model from the multiple models based on the first ratio and the second ratio.

[0089] In one implementation, the acquisition module 510 is used to acquire the second working data of the screw locking working end during the first screw locking process; uniformly interpolate the second working data into multiple interpolation data along the acquisition time of the second working data; uniformly sample the interpolated second working data; and normalize the second working data obtained by uniform sampling to obtain the first working data.

[0090] In one implementation, the determination module 520 is further configured to, when the screw locking state output by the target model is a fault state, perform fault repair according to a correction strategy corresponding to the fault state; wherein, when the fault state is slipped thread locking, the correction strategy is to adjust the locking angle of the screw locking working end, and perform fault repair according to the torque-angle state of the screw locking working end; when the fault state is repeated locking, the correction strategy is to absorb the repeated screws and remove the repeatedly locked screws; when the fault state is oblique locking, the correction strategy is to control the direction, locking angle and torque of the screw locking working end, reversely loosen the screw and adjust the locking angle.

[0091] In one implementation, the acquisition module 510 is also used to obtain the third working data of the screw locking working end during the second screw locking process; perform a clustering operation on the third working data to obtain a cluster data set; analyze the cluster data set, and determine the correspondence between the screw locking state and the first working data based on the analysis results. The correspondence between the screw locking state and the first working data is used to train the target model, and the cluster data set includes the first working data.

[0092] In one implementation, the acquisition module 510 is used to divide the third working data into multiple clusters; repeatedly calculate the distance between the third working data and the class center of each cluster and divide the third data into the cluster corresponding to the class center closest to the third working data until the sum of the squares of the distances between each third working data and each class center does not change, thereby obtaining multiple cluster data sets.

[0093] In one implementation, the acquisition module 510 is used to analyze the cluster data set based on the total number of the third working data, the average distance between each of the third working data in each of the cluster data sets, and the average distance between each of the third working data and the third working data in the closest cluster data set.

[0094] In one implementation, the first working data includes rotation angle data and torque data of the screw locking working end during the first screw locking process.

[0095] The device for determining the screw locking state in the embodiments of the present application can be a device, or a component, integrated circuit, or chip in a terminal. The device can be a mobile electronic device or a non-mobile electronic device. For example, the mobile electronic device can be a mobile phone, a tablet computer, a laptop computer, a PDA, an in-vehicle electronic device, a wearable device, an ultra-mobile personal computer (UMPC), a netbook, or a personal digital assistant (PDA), etc. The non-mobile electronic device can be a server, a network attached storage (NAS), a personal computer (PC), a television (TV), a teller machine, or a self-service machine, etc., and the embodiments of the present application do not specifically limit this.

[0096] The device for determining the screw locking state in the embodiment of the present application may be a device having an operating system. The operating system may be an Android operating system, an iOS operating system, or other possible operating systems, which are not specifically limited in the embodiment of the present application.

[0097] The device for determining the screw locking state provided in the embodiment of the present application can achieve Figures 1 to 4 To avoid repetition, the various processes implemented in the method embodiment are not described here.

[0098] Alternatively, as Figure 6As shown, an embodiment of the present application further provides an electronic device 600, including a processor 601 and a memory 602, wherein the memory 602 stores a program or instruction that can be run on the processor 601, and when the program or instruction is executed by the processor 601, it implements: obtaining the first working data of the screw locking working end in the first screw locking process; inputting the first working data into a target model, and outputting the screw locking state through the target model, wherein the target model includes a correspondence between the first working data and the screw locking state.

[0099] In one implementation, before obtaining the first working data of the screw locking working end in the first screw locking process, in the process of training multiple models, a first number of training data corresponding to the correct screw locking state output by each model is obtained; a first ratio is determined based on the first number and the second number of all training data; in the process of testing multiple models, a third number of test data corresponding to the correct screw locking state output by each model is obtained; a second ratio is determined based on the third number and the fourth number of all test data; and the target model is determined from the multiple models based on the first ratio and the second ratio.

[0100] In one implementation, second working data of the screw locking working end during the first screw locking process is obtained; along the acquisition time of the second working data, the second working data is uniformly interpolated into multiple interpolation data; uniform sampling is performed from the interpolated second working data; and the second working data obtained by uniform sampling is normalized to obtain the first working data.

[0101] In one implementation, after the screw locking state is output through the target model, when the screw locking state output by the target model is a fault state, the fault is repaired according to the correction strategy corresponding to the fault state; wherein, when the fault state is slipped thread locking, the correction strategy is to adjust the locking angle of the screw locking working end, and to repair the fault according to the torque-angle state of the screw locking working end; when the fault state is repeated locking, the correction strategy is to absorb the repeated screws and remove the repeatedly locked screws; when the fault state is oblique locking, the correction strategy is to control the direction, locking angle and torque of the screw locking working end, loosen the screw in the reverse direction and adjust the locking angle.

[0102] In one implementation, before obtaining the first working data of the screw locking working end in the first screw locking process, the third working data of the screw locking working end in the second screw locking process is obtained; the third working data is clustered to obtain a cluster data set; the cluster data set is analyzed, and based on the analysis results, the correspondence between the screw locking state and the first working data is determined, and the correspondence between the screw locking state and the first working data is used to train the target model, and the cluster data set includes the first working data.

[0103] In one implementation, the third working data is divided into multiple clusters; the distance between the third working data and the class center of each cluster is repeatedly calculated and the third data is divided into the cluster corresponding to the class center closest to the third working data until the sum of the squares of the distances between each third working data and each class center does not change, thereby obtaining multiple cluster data sets.

[0104] In one implementation, the cluster data sets are analyzed based on the total number of the third working data, the average distance between each of the third working data in each of the cluster data sets, and the average distance between each of the third working data and the third working data in the closest cluster data set.

[0105] In one implementation, the first working data includes rotation angle data and torque data of the screw locking working end during the first screw locking process.

[0106] The specific execution steps can refer to the various steps of the embodiment of the method for determining the locking state of the screw, and can achieve the same technical effect. To avoid repetition, they will not be described here.

[0107] It should be noted that the electronic devices in the embodiments of the present application include: servers, terminals, or other devices other than terminals.

[0108] The above electronic device structure does not constitute a limitation of the electronic device. The electronic device may include more or fewer components than shown in the figure, or combine certain components, or arrange the components differently. For example, the input unit may include a graphics processing unit (GPU) and a microphone, and the display unit may be configured in the form of a liquid crystal display, an organic light emitting diode, etc. to configure the display panel. The user input unit includes a touch panel and at least one of other input devices. The touch panel is also called a touch screen. Other input devices may include but are not limited to a physical keyboard, function keys (such as volume control buttons, switch buttons, etc.), a trackball, a mouse, and a joystick, which will not be repeated here.

[0109] The memory can be used to store software programs and various data. The memory may mainly include a first storage area for storing programs or instructions and a second storage area for storing data, wherein the first storage area may store an operating system, applications or instructions required for at least one function (such as a sound playback function, an image playback function, etc.), etc. In addition, the memory may include a volatile memory or a non-volatile memory, or the memory may include both volatile and non-volatile memories. Among them, the non-volatile memory may be a read-only memory (ROM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), an electrically erasable programmable read-only memory (EEPROM), or a flash memory. Volatile memory can be random access memory (RAM), static random access memory (SRAM), dynamic random access memory (DRAM), synchronous dynamic random access memory (SDRAM), double data rate synchronous dynamic random access memory (DDRSDRAM), enhanced synchronous dynamic random access memory (ESDRAM), synchronous link dynamic random access memory (SLDRAM) and direct rambus random access memory (DRRAM).

[0110] The processor may include one or more processing units; optionally, the processor may integrate an application processor and a modem processor, wherein the application processor primarily handles operations related to the operating system, user interface, and application programs, and the modem processor primarily processes wireless communication signals, such as a baseband processor. It is understood that the modem processor may not be integrated into the processor.

[0111] An embodiment of the present application also provides a readable storage medium, on which a program or instruction is stored. When the program or instruction is executed by a processor, each process of the embodiment of the method for determining the screw locking state is implemented, and the same technical effect can be achieved. To avoid repetition, it will not be repeated here.

[0112] The processor is the processor in the electronic device described in the above embodiment. The readable storage medium includes a computer-readable storage medium, such as ROM, RAM, magnetic disk or optical disk.

[0113] It should be noted that, in this article, the terms "comprise", "include" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device comprising a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, an element defined by the statement "comprises a ..." does not exclude the presence of other identical elements in the process, method, article or device comprising the element. In addition, it should be noted that the scope of the methods and devices in the embodiments of the present application is not limited to performing functions in the order shown or discussed, and may also include performing functions in a substantially simultaneous manner or in the opposite order according to the functions involved. For example, the described method may be performed in an order different from that described, and various steps may also be added, omitted, or combined. In addition, the features described with reference to certain examples may be combined in other examples.

[0114] Through the description of the above implementation methods, those skilled in the art can clearly understand that the above-mentioned embodiment methods can be implemented by means of software plus the necessary general hardware platform, and of course can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of the present application is essentially or the part that contributes to the prior art can be embodied in the form of a computer software product, which is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk), including a number of instructions for enabling a terminal (which can be a mobile phone, computer, server, or network device, etc.) to execute the methods described in each embodiment of the present application.

[0115] The embodiments of the present application are described above in conjunction with the accompanying drawings, but the present application is not limited to the above-mentioned specific implementation methods. The above-mentioned specific implementation methods are merely illustrative and not restrictive. Under the guidance of this application, ordinary technicians in this field can also make many forms without departing from the purpose of this application and the scope of protection of the claims, all of which are within the protection of this application.

Claims

1. A method for determining the locking state of a screw, characterized in that: include: Acquire first working data of the screw locking working end during the first screw locking process; The first working data is input into a target model, and a screw locking state is output through the target model. The target model includes a corresponding relationship between the first working data and the screw locking state.

2. The determination method according to claim 1, characterized in that Before obtaining the first working data of the screw locking working end in the first screw locking process, the method further includes: During the process of training the multiple models, obtaining a first quantity of training data corresponding to the correct screw locking state output by each of the models; determining a first ratio based on the first quantity and a second quantity of all training data; In the process of testing the plurality of models, obtaining a third quantity of test data corresponding to a correct screw locking state output by each model; determining a second ratio based on the third quantity and a fourth quantity of all test data; The target model is determined from the plurality of models according to the first ratio and the second ratio.

3. The determination method according to claim 1, characterized in that The obtaining of first working data of the screw locking working end during the first screw locking process includes: Acquire second working data of the screw locking working end during the first screw locking process; uniformly interpolating the second working data into a plurality of interpolation data along an acquisition time of the second working data; uniformly sampling from the interpolated second working data; Normalizing the second working data obtained by uniform sampling to obtain the first working data.

4. The determination method according to claim 1, characterized in that After outputting the screw locking state through the target model, the method further includes: When the screw locking state output by the target model is a fault state, performing fault repair according to a correction strategy corresponding to the fault state; Wherein, when the fault state is slipped thread locking, the correction strategy is to adjust the locking angle of the screw locking working end and perform fault repair according to the torque-angle state of the screw locking working end; In the case where the fault state is repeated locking, the correction strategy is to suck out the repeated screws and remove the repeatedly locked screws; In the case where the fault state is oblique locking, the correction strategy is to control the direction, locking angle and torque of the screw locking working end, loosen the screw in the reverse direction and adjust the locking angle.

5. The determination method according to claim 1, characterized in that: Before obtaining the first working data of the screw locking working end in the first screw locking process, the method further includes: Acquiring third working data of the screw locking working end during the second screw locking process; performing a clustering operation on the third working data to obtain a clustered data set; The cluster data set is analyzed, and based on the analysis results, the correspondence between the screw locking state and the first working data is determined, and the correspondence between the screw locking state and the first working data is used to train the target model, and the cluster data set includes the first working data.

6. The determination method according to claim 5, characterized in that: The clustering operation is performed on the third working data to obtain a cluster data set, including: dividing the third working data into a plurality of clusters; Repeatedly calculate the distance between the third working data and the class center of each cluster and divide the third data into the cluster corresponding to the class center closest to it until the sum of the squares of the distances between each third working data and each class center does not change, thereby obtaining multiple cluster data sets.

7. The determination method according to claim 5, characterized in that: The analyzing the cluster data set includes: The cluster data sets are analyzed according to the total number of the third working data, the average distance between the third working data in each cluster data set, and the average distance between each third working data and the third working data in the closest cluster data set.

8. The determination method according to claim 1, characterized in that: The first working data includes rotation angle data and torque data of the screw locking working end during the first screw locking process.

9. A device for determining the locking state of a screw, characterized in that: include: An acquisition module, configured to acquire first working data of the screw locking working end during a first screw locking process; The determination module is configured to input the first working data into a target model and output a screw locking state through the target model, wherein the target model includes a correspondence between the first working data and the screw locking state.

10. An electronic device, characterized in that: The invention comprises a processor, a memory, and a program or instruction stored in the memory and executable on the processor, wherein when the program or instruction is executed by the processor, the steps of the method for determining the screw locking state according to any one of claims 1 to 8 are implemented.

11. A readable storage medium, characterized in that: The readable storage medium stores a program or instruction, and when the program or instruction is executed by the processor, the steps of the method for determining the screw locking state according to any one of claims 1 to 8 are implemented.

Images