Screw Feeding Pressure Adaptive Adjustment Method, Device and Equipment

Abstract

The present invention relates to a method, device and equipment for self - adaptive adjustment of screw feeding pressure. The method includes: constructing a Lagrangian control model for the screw feeding equipment to obtain a control reference parameter set; acquiring real - time air pressure data stream and calculating an air pressure regulation instruction; constructing a torque - rotation speed - temperature mapping function and calculating a torque output parameter; calculating a torque compensation parameter and performing an integral operation with the data of the locking angle sensor to generate a torque - locking angle characteristic curve; performing a locking state evaluation to obtain a state evaluation index, performing a deviation analysis on the state evaluation index and a set locking threshold, and obtaining a system correction matrix through multi - dimensional parameter collaborative calculation; inputting the system correction matrix into the Lagrangian control model, updating the dynamic weight coefficient, and outputting an optimized control instruction set. The implementation of the present invention improves the reliability and maintenance efficiency of the equipment, and solves the technical problems of traditional screw feeding equipment in dynamic control, parameter optimization and fault warning.

Description

Technical Field

[0001] The present invention relates to the technical field of screw feeding, and particularly relates to a method, device and equipment for self-adaptive adjustment of screw feeding pressure. Background Art

[0002] Traditional screw feeding equipment mainly relies on fixed air pressure parameters and preset torque values for control, and cannot dynamically adjust according to the characteristics of different specifications of screws, easily resulting in problems such as unstable feeding, jamming, and incomplete locking.

[0003] Currently, screw feeding equipment on the market generally has technical bottlenecks such as difficult to accurately control feeding parameters, large torque output fluctuations, and insufficient temperature compensation. Especially in the assembly process of high-precision electronic products, due to the diversification of screw specifications and complex assembly working conditions, the traditional open-loop control method is difficult to meet the requirements of efficient and stable assembly, affecting product assembly quality and production efficiency. In addition, there is a lack of a collaborative control mechanism for screw feeding pressure and electric screwdriver torque output in the prior art, and it is impossible to achieve dynamic optimization of the feeding process and the locking process. At the same time, due to the lack of effective state monitoring and fault warning means, abnormal conditions during the operation of the equipment are difficult to detect and handle in a timely manner, increasing the production maintenance cost and the difficulty of quality control. Summary of the Invention

[0004] The main purpose of the present invention is to provide a method, device and equipment for self-adaptive adjustment of screw feeding pressure, which improves the reliability and maintenance efficiency of the equipment and solves the technical problems of traditional screw feeding equipment in dynamic control, parameter optimization and fault warning.

[0005] To achieve the above purpose, the present invention provides a method for self-adaptive adjustment of screw feeding pressure, including the following steps:

[0006] Extract screw specification parameters of the screw feeding equipment, and collect the electric screwdriver rotation speed signal and the initial torque signal to construct a Lagrangian control model, and obtain a control reference parameter group;

[0007] Obtain the real-time air pressure data stream, and calculate the conveying cycle deviation and the screw impact stress value according to the control reference parameter group, and obtain the air pressure regulation instruction through differential compensation operation;

[0008] Transmit the air pressure regulation instruction to the electric screwdriver driver, construct a torque-rotation speed-temperature mapping function in combination with the real-time displacement data collected by the angular displacement sensor, and calculate the torque output parameter through the least squares method;

[0009] Perform self-adaptive compensation calculation on the torque output parameter to obtain a torque compensation parameter, and perform integral operation on the torque compensation parameter and the locking angle sensor data to generate a torque-locking angle characteristic curve;

[0010] Based on the torque-locking angle characteristic curve, the locking state is evaluated to obtain a state evaluation index. The deviation analysis is performed between the state evaluation index and a set locking threshold, and a system correction matrix is obtained through collaborative calculation of multi-dimensional parameters;

[0011] The system correction matrix is input into the Lagrangian control model to update the dynamic weight coefficient, and an optimized control instruction set is output.

[0012] The present invention also provides a device for adaptively adjusting the screw feeding pressure, including:

[0013] An acquisition module, configured to extract screw specification parameters of a screw feeding device, and acquire an electric screwdriver rotation speed signal and an initial torque signal to construct a Lagrangian control model, so as to obtain a set of control reference parameters;

[0014] A calculation module, configured to obtain a real-time air pressure data stream, and calculate a conveying cycle deviation and a screw impact stress value according to the set of control reference parameters, and obtain an air pressure regulation instruction through difference compensation operation;

[0015] A construction module, configured to transmit the air pressure regulation instruction to an electric screwdriver driver, jointly acquire real-time displacement data collected by an angular displacement sensor to construct a torque-rotation speed-temperature mapping function, and calculate torque output parameters through the least squares method;

[0016] A compensation module, configured to perform adaptive compensation calculation on the torque output parameters to obtain torque compensation parameters, and perform integral operation on the torque compensation parameters and the data of a locking angle sensor to generate a torque-locking angle characteristic curve;

[0017] An analysis module, configured to perform locking state evaluation based on the torque-locking angle characteristic curve to obtain a state evaluation index, perform deviation analysis between the state evaluation index and a set locking threshold, and obtain a system correction matrix through collaborative calculation of multi-dimensional parameters;

[0018] An output module, configured to input the system correction matrix into the Lagrangian control model to update the dynamic weight coefficient, and output an optimized control instruction set.

[0019] The present invention also provides a computer device, including a memory and a processor. A computer program is stored in the memory, and when the processor executes the computer program, the steps of the method described in any one of the above are implemented.

[0020] The present invention also provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the steps of the method described in any one of the above are implemented.

[0021] In summary, the technical solution provided by the present invention realizes the adaptive adjustment of the screw feeding pressure by establishing a Lagrangian control model, integrates multiple technical features such as pneumatic control, torque compensation, and temperature correction, and forms a complete closed-loop control system. The present invention adopts the multi-sensor data fusion technology to collect and process parameters such as air pressure, displacement, and temperature in real time, establishes an accurate torque-locking angle characteristic curve, and ensures the stability of the feeding and locking processes through the dynamic weight update mechanism; at the same time, the present invention introduces a fault alarm and status evaluation mechanism, discovers and processes abnormal conditions in a timely manner through multi-dimensional parameter collaborative calculation, significantly improves the reliability and maintenance efficiency of the equipment, and solves the technical problems of traditional screw feeding equipment in dynamic control, parameter optimization, and fault warning. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] Figure 1 is a schematic diagram of the steps of the screw feeding pressure adaptive adjustment method in an embodiment of the present invention;

[0023] Figure 2 is a structural block diagram of the screw feeding pressure adaptive adjustment device in an embodiment of the present invention;

[0024] Figure 3 is a structural schematic block diagram of a computer device in an embodiment of the present invention.

[0025] The realization, functional characteristics, and advantages of the object of the present invention will be further described in conjunction with the embodiments with reference to the drawings. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0026] In order to make the object, technical solution, and advantages of the present invention clearer, the present invention will be further described in detail below with reference to the drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0027] Referring to Figure 1 , this embodiment provides a screw feeding pressure adaptive adjustment method, including the following steps:

[0028] S1, extract the screw specification parameters of the screw feeding equipment, and collect the electric screwdriver rotation speed signal and the initial torque signal to construct a Lagrangian control model, and obtain a control reference parameter group;

[0029] Among them, the screw specification parameters are entered through the input interface of the liquid crystal screen control system. The specification parameters include the total length, diameter, weight, material, etc. of the screw. The entered screw parameters are classified according to the preset specification range to obtain the initial matrix of screw parameters, which contains the characteristic data of all entered screws and is used for subsequent analysis and calculation. The ratio of the total length L and the nut diameter D of the screws in the initial matrix of screw parameters is calculated to screen out the screw specifications that meet the requirements. Only the effective parameter groups with L / D greater than 1.3 are selected to ensure that the selected screws are geometrically suitable for the operation of the feeding device. At the same time, the rotational speed signal of the electric screwdriver in the no-load state is collected by the encoder. These analog signals are digitally processed, and the corresponding rotational speed digital quantity is obtained through signal sampling and conversion. In order to enhance the dynamic response ability of the control model, the angular displacement signal during the movement of the electric screwdriver is collected. The angular displacement sensor monitors the displacement change of the electric screwdriver in real time, and through differential operation on the displacement data, the real-time rotational speed change rate of the electric screwdriver is obtained, reflecting the speed change of the electric screwdriver in different working stages, which helps to accurately describe the dynamic characteristics of the system during the screw locking process. During the screw locking process, since the increase in temperature will affect the performance and torque output of the electric screwdriver, a temperature sensor is introduced to collect the temperature of the environment and the equipment in real time. The temperature data is input into the compensation function to calculate the initial torque coefficient, and the initial torque coefficient is temperature-corrected according to the current temperature to obtain the corrected torque coefficient. Based on the effective parameter group, the rotational speed digital quantity and the rotational speed change rate, the motion equation of the electric screwdriver torque output system is established. This motion equation describes the dynamic characteristics of the electric screwdriver during the locking process and at the same time reflects the state of screw feeding. In order to accurately adjust the relationship between the air pressure and the rotational speed during the feeding process, the rotational speed and pressure parameters of the feeder in the compressed air pipeline are sampled, and the functional relationship between the air pressure and the rotational speed is obtained through data fitting, describing the rotational speed change of the feeding system under different air pressure conditions, which helps to achieve accurate control of the air pressure. The motion equation of the electric screwdriver torque output system, the corrected torque coefficient, and the air pressure-rotational speed functional relationship are substituted into the Lagrangian control model to construct a complete dynamic model, and the control reference parameter group is obtained. The dynamic model can comprehensively describe the dynamic characteristics of the screw feeding and locking process, and the control reference parameter group provides an important basis for the adaptive adjustment of the feeding pressure.

[0030] S2, obtain the real-time air pressure data stream, calculate the conveying cycle deviation and the screw impact stress value according to the control reference parameter group, and obtain the air pressure regulation command through the difference compensation operation;

[0031] Specifically, air pressure data is collected by distributed pressure sensors installed in the compressed air supply pipeline. These distributed pressure sensors can sense the air pressure changes in the pipeline in real time. The analog signals collected are converted into digital signals through a signal conversion module to obtain the initial air pressure data. The initial data is processed by a digital filter to eliminate noise interference in the environment. After filtering, a real-time air pressure data stream is obtained. Based on the feeder rotation speed parameter in the control reference parameter group, a periodic analysis is performed on the real-time air pressure data stream. By dynamically monitoring the feeding process, the time required for a single screw to be transported from the feeder to the electric screwdriver chuck is calculated, that is, the transport time value of a single screw. Based on the transport time value, the feeding cycle is calculated, and the obtained actual transport cycle is compared with the preset standard cycle of 0.6 seconds to obtain the deviation value of the transport cycle. If there is a significant deviation between the transport cycle and the standard cycle, it will affect the accuracy and efficiency of screw locking, and these deviations are corrected in real time. At the same time, the impact force when the screw reaches the electric screwdriver chuck is detected by a high-precision torque sensor to obtain the stress condition of the screw during transportation. Through stress conversion, the impact stress value of the screw is obtained. A threshold judgment is made on the impact stress value of the screw, and the actual impact stress is compared with the preset stress threshold to determine whether the screw is subjected to excessive impact or extrusion during transportation. If the impact stress exceeds the preset threshold, it means that the air pressure during the feeding process needs to be adjusted to reduce the stress of the screw during transportation and ensure the smoothness of screw transportation. To calculate the required stress compensation amount, based on the air pressure function in the control reference parameter group and combined with the current impact stress value, an accurate stress compensation amount is calculated. The transport cycle deviation and the stress compensation amount are input into the adaptive algorithm module. By performing weighted calculations on these two parameters, an air pressure correction value for correcting the current air pressure is obtained. The adaptive algorithm module dynamically adjusts the air pressure according to the weights of different deviation values to ensure that the feeding process can proceed smoothly within the standard cycle. The difference between the air pressure correction value and the current air pressure data is used as the input, and an air pressure control command is generated through compensation operation. The air pressure control command is used to adjust the supply amount of compressed air in real time to optimize the feeding state of the screw, ensuring that sufficient pressure can be provided to quickly transport the screw during the feeding process without damaging the screw due to excessive pressure.

[0032] S3, transmit the air pressure control command to the electric screwdriver driver, construct a torque-rotation speed-temperature mapping function in combination with the real-time displacement data collected by the angular displacement sensor, and calculate the torque output parameter through the least squares method;

[0033] It should be noted that the air pressure control instruction is input into the electric screwdriver driver, and the air pressure data is segmented according to the preset pressure range to obtain the control reference value of the electric screwdriver. The control reference value is used to guide the driving operation of the electric screwdriver at different air pressure levels, ensuring that the air pressure regulation during the feeding process can adapt to the different requirements of the screw locking process. Through segmentation, the electric screwdriver can perform refined driving control for different pressure levels to improve its dynamic response ability. The rotation state of the chuck is detected in real time by an angular displacement sensor installed on the electric screwdriver chuck. These detection data can intuitively reflect the operating conditions of the electric screwdriver. The original signal obtained by the angular displacement sensor is processed by a signal conditioning circuit to remove noise and interference, obtaining high-quality real-time displacement data. These data represent the current displacement state of the electric screwdriver, and through time series analysis and differential calculation, the real-time rotation speed value and acceleration value of the electric screwdriver are obtained. The real-time rotation speed and acceleration are important parameters describing the dynamic characteristics of the electric screwdriver, jointly constituting the basic dynamic characteristics of the electric screwdriver's movement, which helps to describe the dynamic response of the electric screwdriver under different pressure conditions. At the same time, the temperature in the working environment of the electric screwdriver is collected by a temperature sensor. The change in temperature will affect the movement characteristics and torque output of the electric screwdriver. The ambient temperature data collected by the temperature sensor is input into the compensation calculation unit, and the temperature compensation factor is obtained through correction calculation. The temperature compensation factor is used to correct the performance deviation caused by the change in ambient temperature, ensuring that the working state of the electric screwdriver can remain consistent under different temperature conditions. The compensation calculation is designed based on the influence of temperature on material properties and mechanical movement, and the corrected parameters more accurately reflect the actual working conditions of the electric screwdriver. On this basis, based on the control reference value and real-time rotation speed value of the electric screwdriver, a speed response equation is constructed. This equation can describe the response characteristics of the electric screwdriver under different pressure and rotation speed conditions, and combined with the temperature compensation factor, substituting it into the speed response equation to obtain the corrected rotation speed curve. According to the corrected rotation speed curve and real-time acceleration value, a dynamic characteristic equation of the electric screwdriver is established. This equation can comprehensively describe the relationship between the speed, acceleration and external conditions of the electric screwdriver. Through function transformation of this equation, a mapping function including the relationship between torque, rotation speed and temperature is obtained, that is, the torque-rotation speed-temperature mapping function. This mapping function is used to describe the torque characteristics output by the electric screwdriver under different temperature and rotation speed conditions, and is the core basis for torque control in the feeding pressure regulation. In order to obtain specific torque output parameters, the least squares method is used to calculate the constructed torque-rotation speed-temperature mapping function. The least squares method is an error minimization method. By fitting the collected data, the function form that best conforms to the data characteristics is found. Through error iterative calculation, an initial torque value sequence is obtained. This sequence represents the initial torque output situation of the electric screwdriver under different working conditions. The weighted average calculation is performed on the initial torque value sequence. Considering the importance difference of the initial torque values under different working conditions, higher weight assignment is given to key data to obtain more stable and accurate torque output parameters.This torque output parameter is used for the real-time control of the electric screwdriver to ensure that the electric screwdriver can provide appropriate torque during the screw locking process to achieve stable locking of the screw.

[0034] S4. Perform an adaptive compensation calculation on the torque output parameter to obtain a torque compensation parameter, and perform an integration operation on the torque compensation parameter and the data of the locking angle sensor to generate a torque-locking angle characteristic curve.

[0035] Specifically, sample and analyze the torque output parameter, perform data calibration based on the working range of the electric screwdriver torque to obtain a standardized torque reference sequence. This torque reference sequence serves as the basis for subsequent compensation calculations to ensure the accuracy and consistency of torque output. To improve the compensation process and ensure that the torque output matches the actual requirements, by constructing a compensation calculation model, substitute the rotational speed data of the electric screwdriver in the no-load state into the compensation calculation model for comprehensive analysis and calculation to obtain a torque compensation parameter. This compensation parameter can effectively correct the torque deviation caused by system nonlinearity or external interference and improve the torque accuracy of the entire system during the actual locking process. Dynamically compensate the torque accuracy of the electric screwdriver according to the torque compensation parameter. During the compensation process, obtain the compensated torque value through data correction operations. The compensated torque value more accurately reflects the actual torque output of the electric screwdriver during the locking process, ensuring that the torque can adapt to the mechanical requirements of screw locking in real time and avoiding problems such as locking failure or damage caused by insufficient or excessive torque. At the same time, the locking angle sensor collects the change data of the angle during the locking process and samples these data. Since there is noise and invalid signals in the angle data, obtain effective angle values through data screening. These effective angle values represent the actual rotation of the screw during the locking process. Match the compensated torque value with the effective angle values in time series, and obtain torque-angle data pairs through data alignment operations, reflecting the torque and the corresponding rotational angle changes at each instant during the locking process. To analyze the dynamic behavior of the screw during the locking process, classify the torque-angle data pairs according to the thread type of the screw. Different types of threads exhibit different torque and angle change laws during the locking process. Classifying them helps to more accurately describe the locking characteristics. Segment the data of each type of thread after classification to obtain several characteristic intervals. Each characteristic interval represents a certain stage during the locking process, such as the initial locking stage, the intermediate transition stage, and the final tightening stage. The mechanical behaviors of each stage are different. Perform an integration calculation on the data within the characteristic interval, and obtain the parameters of the characteristic curve through cumulative operations. The integration operation describes the overall behavior of the screw during the locking process by accumulating the changes in torque and angle, and the integration result reflects the cumulative effect of torque and angle over time. Based on the characteristic curve parameters, use the method of function fitting to reconstruct the curve and generate a torque-locking angle characteristic curve to describe the relationship between the torque and the angle of the screw during the locking process.

[0036] S5. Based on the torque-locking angle characteristic curve, evaluate the locking state to obtain a state evaluation index, perform deviation analysis on the state evaluation index and the set locking threshold, and obtain a system correction matrix through multi-dimensional parameter collaborative calculation;

[0037] Among them, data analysis is carried out based on the torque-locking angle characteristic curve. By analyzing the characteristic curves at different stages of the screw assembly process, key nodes are extracted to obtain the locking state sequence describing the entire locking process. These key nodes represent important events or state changes during the locking process, such as initial locking, tightening the screw to a predetermined torque, final tightening, etc. Based on the locking state sequence, the accuracy of the torque is monitored online. By analyzing the torque range, torque state data reflecting the locking state are obtained. These data are used to describe the stability and change trend of the torque output during the locking process, so as to evaluate the quality and reliability of the locking. The torque state data are input into the evaluation unit to conduct a comprehensive state evaluation of the screw locking process. The evaluation unit monitors the abnormalities during the locking process in real time based on the fault alarm and counting functions, and generates state evaluation indicators. During the evaluation process, three dimensions are monitored, namely material shortage, material jamming, and torque abnormality. Material shortage means that the screw fails to be supplied in time, resulting in the inability to continue the locking process; material jamming indicates that the screw is blocked or jammed during the feeding process, affecting the normal locking process; while torque abnormality may indicate that the electric screwdriver fails to output the appropriate torque during the locking process, all of which will lead to unqualified final assembly quality. By monitoring these abnormalities and based on the threshold detection method, an abnormality identification matrix is generated to identify and record all abnormal situations during the locking process. The state evaluation indicators and the abnormality identification matrix are subjected to data fusion, and the comprehensive state score of the entire locking process is calculated through weighted operations. Different weights are assigned to different abnormalities according to their severity, reflecting the overall impact of each abnormality on the locking quality. The comprehensive state score is a quantitative evaluation of the overall quality of the locking process, indicating the quality of the current locking state. The comprehensive state score is compared with the preset locking threshold, and a deviation parameter group is obtained through difference calculation. If the comprehensive score deviates from the set locking threshold, it means that there are certain abnormalities in the current locking process, and further adjustments are needed to correct these deviations. Based on the deviation parameter group, a multi-dimensional state space is constructed to conduct collaborative analysis on various deviations existing in the locking process. The multi-dimensional state space contains various state variables, such as torque, rotational speed, angle, temperature, etc. These variables jointly describe the dynamic characteristics of the locking process. Through collaborative analysis in the multi-dimensional state space, the mutual influence between various state variables is accurately found, and the corresponding parameter correction amounts are calculated. These parameter correction amounts are used to dynamically adjust the working state of the system to eliminate abnormal phenomena during the locking process and ensure that the entire process proceeds within a predetermined range. The parameter correction amounts are subjected to dimensionality reduction processing according to the dynamic model. Through dimensionality reduction, the calculation process is simplified, redundant information is removed, and more concise and effective adjustment parameters are obtained. After the dimensionality reduction processing is completed, a system correction matrix is obtained through matrix transformation. The system correction matrix is used to globally correct and optimize each parameter in the control system to further improve the stability and reliability of the locking process.Each element in the correction matrix represents the adjustment amplitude of different state variables. By dynamically adjusting these variables, the feeding system realizes the real-time adaptive control of multiple parameters such as torque, pressure, and rotational speed, thereby improving the accuracy and consistency of the screw locking process.

[0038] S6. Input the system correction matrix into the Lagrangian control model, update the dynamic weight coefficient, and output the optimized control instruction set.

[0039] Specifically, substitute the system correction matrix into the motion equation of the electric screwdriver torque output system and the air pressure-rotation speed function relationship for matrix operations. Through this process, combined with the correction matrix and the existing dynamic model, the initial weight data is obtained. The initial weight data reflects the key parameter relationships of the system in the current state. The substitution of the system correction matrix and matrix operations enable these initial weight data to reflect the actual dynamic characteristics of the current feeding system. Conduct a feature analysis on the initial weight data. By analyzing the characteristics of the initial weights, the parameters that play a key role in the system dynamics can be identified. Combine with the control reference parameter group to optimize the weight data and obtain the updated coefficient matrix. The updated coefficient matrix corrects the initial weight data in multiple optimization operations to make the dynamic response characteristics of the system more accurate and meet the predetermined control requirements. Perform a weight iteration operation on the updated coefficient matrix. Through repeated adjustment and iteration, the dynamic weight coefficient is obtained. The update and iteration of the weight coefficient aim to optimize the overall control performance of the system, enabling the feeding and locking processes to quickly respond to external changes and achieve a stable operation effect. To ensure that the dynamic weight coefficient can adapt to the changes in the actual feeding and locking processes, the weight coefficient is corrected online according to the rotation speed and pressure parameters of the feeder, and the optimized control parameter group is obtained. The rotation speed and pressure parameters of the feeder have a direct impact on the feeding effect of the screws and the final locking quality. Through online correction, it is ensured that the control parameter group is always consistent with the actual operating state of the feeder, improving the dynamic response ability of the system. Substitute the optimized control parameter group into the Lagrange control model. Combine with the previously obtained parameter relationships to obtain the control variable sequence through parameter mapping. The control variable sequence is used to describe the states of each control link in the feeding and locking processes, ensuring that the control variables of each operation change within the optimal range. Perform an analysis and processing on the control variable sequence and match the parameters in combination with the torque-locking angle characteristic curve. Through this process, it is ensured that the control variables can reflect the characteristics of the actual locking process, and the reference instruction set is obtained. The reference instruction set is the basis of the entire control system, including the operation instructions for each step of feeding and locking. Divide the reference instruction set according to the electric screwdriver control and feeding control. The electric screwdriver control instructions are used to manage the torque output, rotation speed adjustment, and angle control of the electric screwdriver, ensuring that the screws are tightened step by step according to the set path and torque; the feeding control instructions are used to control the supply of compressed air, the rotation speed adjustment of the feeder, and the conveying of screws, ensuring that the screws can reach the locking position accurately at the required moment. Through instruction parsing, the control tasks are decomposed into independent sub-control instructions, making the entire control process more refined and modular, facilitating precise adjustment and real-time response. Dynamically combine the sub-control instructions, integrate all the instructions through coordinated operations, and output the optimized control instruction set.The purpose of the coordinated operation is to achieve coordination between the electric screwdriver control and the feeding control, so that the feeding and locking operations can be carried out synchronously, avoiding problems such as untimely feeding or insufficient locking torque, thereby ensuring the stability and efficiency of the system operation. The optimized control instruction set comprehensively considers the current state of the system, environmental changes, and real-time data feedback during the locking process, and can achieve optimal control of the feeding and locking processes under different working conditions.

[0040] In one example, the screw specification parameters of the screw feeding device are extracted, and the electric screwdriver rotation speed signal and the initial torque signal are collected to construct a Lagrangian control model, obtaining a control reference parameter group, including:

[0041] The screw specification parameters are entered through the input interface of the LCD screen control system, and the data is classified according to the preset specification range to obtain the initial screw parameter matrix;

[0042] The ratio of the total length L to the nut diameter D of the screws in the initial screw parameter matrix is calculated, and the effective parameter groups with L / D greater than 1.3 are screened out. At the same time, the no-load rotation speed signal of the electric screwdriver collected by the encoder is digitally processed, and the rotation speed digital quantity is obtained through signal sampling conversion;

[0043] The displacement data collected by the angular displacement sensor is subjected to differential operation to obtain the real-time rotation speed change rate, and the temperature data collected by the temperature sensor is input into the compensation function to obtain the initial torque coefficient, and the initial torque coefficient is corrected for temperature to obtain the corrected torque coefficient;

[0044] Based on the effective parameter group, the rotation speed digital quantity, and the rotation speed change rate, the motion equation of the electric screwdriver torque output system is established, and the rotation speed and pressure parameters of the feeder in the compressed air pipeline are sampled, and the air pressure-rotation speed function relationship is obtained through data fitting;

[0045] The motion equation of the electric screwdriver torque output system, the corrected torque coefficient, and the air pressure-rotation speed function relationship are substituted into the Lagrangian control model to construct a dynamic model, obtaining a control reference parameter group.

[0046] In this example, the screw specification parameters are entered through the input interface of the LCD screen control system. The LCD screen control system is the interaction interface between the operator and the feeding system. The operator intuitively inputs various screw parameters through the LCD screen, such as the total length of the screw 、the diameter of the nut 、thread type, material, and quality, etc. After these parameters are entered, the screws are classified according to the preset specification range, and a data set containing all screw characteristics is constructed, that is, the initial screw parameter matrix. The initial matrix is expressed as:

[0047] ;

[0048] Among them, represents the total length of the th screw, represents the diameter of the nut, is the thread type, is the screw material, is the screw quality. The initial matrix contains the basic specification information of all screws. For the total length of the screws and the nut diameter in the initial matrix of screw parameters, a ratio calculation is performed to screen out the effective parameter groups suitable for the feeding device. The ratio calculation is completed through the following formula:

[0049] ;

[0050] For all screws, if the ratio , then these screws are regarded as effective screw parameter groups, denoted as the effective matrix . At the same time, the rotational speed signal of the electric screwdriver in the no-load state is collected through an encoder. These rotational speed signals are initially analog signals. Through signal sampling and conversion, the no-load rotational speed signals collected by the encoder are converted into digital quantities, denoted as . At the same time, the angular displacement sensor collects the chuck displacement of the electric screwdriver in real time to obtain the displacement data . Through the first-order differential operation on the displacement data, the real-time rotational speed change rate of the electric screwdriver is obtained, that is:

[0051] ;

[0052] Among them, represents the real-time rotational speed, reflecting the speed change of the system at different time points. These data are used to describe the dynamic behavior of the electric screwdriver and provide a basis for establishing the motion equation. In order to consider the influence of temperature on the system performance, a temperature sensor is introduced to collect the ambient temperature data, denoted as . The temperature data is input into the compensation function for correction. The compensation function is used to calculate the initial torque coefficient . Considering the influence of temperature on the material properties and equipment performance, the initial torque coefficient needs to be corrected by temperature. The corrected torque coefficient is expressed as:

[0053] ;

[0054] Among them, is the corrected torque coefficient, is the temperature correction coefficient, is the reference temperature value. This correction takes into account the influence of temperature increase or decrease on the torque output, so as to ensure that the torque output remains stable in different temperature environments. Based on the effective parameter group , rotational speed digital quantity and real-time rotational speed change rate , to establish the motion equation of the electric screwdriver torque output system. The dynamic behavior of the electric screwdriver is described by the following equation:

[0055] ;

[0056] Among them, represents the moment of inertia of the electric screwdriver, represents the damping coefficient, represents the load torque generated during the locking process due to friction between the screw and the material, etc. Through this equation, the dynamic response of the electric screwdriver when applying torque is described. At the same time, the rotational speed and pressure parameters of the feeder in the compressed air pipeline are sampled to understand the response of the system under different air pressure conditions. By fitting these data, the functional relationship between air pressure and rotational speed is obtained, denoted as:

[0057] ;

[0058] Among them, represents the rotational speed of the feeder under pressure , , , are fitting coefficients. This functional relationship describes the influence of compressed air on the rotational speed of the feeder, which helps to achieve precise control of the feeding process. Substitute the motion equation of the electric screwdriver torque output system, the corrected torque coefficient, and the air pressure - rotational speed functional relationship into the Lagrangian control model to construct the dynamic model of the entire system. The Lagrangian control model is used to describe the energy conversion and motion characteristics of the system. Through the Lagrangian equation, a set of control reference parameters is obtained to guide the subsequent control strategy. The Lagrangian equation is expressed as:

[0059] ;

[0060] Among them, is the Lagrangian function, defined as the difference between the kinetic energy of the system and the potential energy , that is and represent the generalized coordinate and generalized velocity respectively, represents the generalized force. In this system, the generalized coordinate is represented by the displacement, rotation angle, etc. of the screw, and the generalized velocity is the rotational speed of the electric screwdriver. Through the above equations, the motion equation, corrected torque, and air pressure function are combined to obtain a set of control reference parameters for the entire dynamic system. These control reference parameters include a series of variables such as the rotational speed, torque output, and air pressure of the electric screwdriver.

[0061] In one example, a real-time air pressure data stream is obtained, and the conveying cycle deviation and the screw impact stress value are calculated according to a control reference parameter group. An air pressure regulation instruction is obtained through difference compensation operation, including:

[0062] Data acquisition is performed by distributed pressure sensors arranged in the compressed air supply pipeline. Initial air pressure data is obtained through signal conversion, and the initial air pressure data is input into a digital filter to eliminate environmental noise interference, obtaining a real-time air pressure data stream;

[0063] According to the feeder rotation speed parameter in the control reference parameter group, periodic analysis is performed on the real-time air pressure data stream to obtain the conveying time value of a single screw, and periodic calculation is performed based on the conveying time value of a single screw. The calculation result is compared with a preset standard cycle of 0.6 seconds to obtain the conveying cycle deviation;

[0064] The force when the screw reaches the electric screwdriver chuck is detected by a high-precision torque sensor, the screw impact stress value is obtained through stress conversion, and a threshold judgment is made on the screw impact stress value. The stress compensation amount is calculated based on the air pressure function in the control reference parameter group;

[0065] The conveying cycle deviation and the stress compensation amount are input into an adaptive algorithm module, and an air pressure correction value is obtained through weighted calculation. Based on the difference between the air pressure correction value and the current air pressure data, an air pressure regulation instruction is generated through compensation operation.

[0066] In this example, multiple pressure sensors are reasonably arranged in the supply pipeline. These sensors are used to capture the pressure change of the compressed air in the supply pipeline to obtain the original air pressure signal, denoted as . Since these signals are interfered by environmental noise during the acquisition process, such as mechanical vibration, pipeline resonance, etc., the collected initial air pressure data is digitally processed through a signal conversion module and input into a digital filter to eliminate environmental noise interference, obtaining a real-time air pressure data stream after filtering processing . Periodic analysis is performed on the real-time air pressure data according to the feeder rotation speed parameter in the control reference parameter group. The rotation speed of the feeder is an important factor determining the screw conveying speed, and the change in air pressure directly affects the rotation speed of the feeder. By performing periodic analysis, the time required for a single screw to be conveyed from the feeder to the target position is obtained, denoted as the conveying time . This time value is obtained by observing the air pressure fluctuation cycle because each time a screw is conveyed in place, it will cause a characteristic fluctuation in the air pressure. The conveying time of a single screw is calculated by the following formula:

[0067] ;

[0068] where Indicates the frequency of air pressure fluctuations during the feeding process. After obtaining the conveying time, cycle calculation is performed based on this time, and the calculation result is compared with the preset standard cycle seconds to obtain the deviation of the conveying cycle

[0069] ;

[0070] If there is a deviation between the conveying time and the standard time, it indicates that there is an abnormality in the current feeding process. For example, insufficient air pressure causes the feeding speed to slow down, or excessive air pressure causes the screws to be conveyed too quickly. At the same time, in order to ensure the stability of the screws when they are conveyed to the electric screwdriver chuck, the force exerted by the screws when they reach the electric screwdriver chuck is detected by a high-precision torque sensor. The magnitude of this force can reflect the impact intensity of the screws during the conveying process. Through stress conversion, the impact stress value generated by the screws when they hit the chuck is obtained and denoted as . The stress conversion formula is:

[0071] ;

[0072] Among them, represents the impact force of the screws when they reach the chuck, represents the cross-sectional area of contact between the screws and the chuck. Through calculation, the impact intensity of the screws when they are conveyed in place is accurately evaluated. Threshold judgment is performed on the screw impact stress value, and is compared with the set safety stress threshold . If , it indicates that the impact intensity of the screws is too large, which will affect the integrity of the screws or the reliability of locking. Based on the air pressure function in the control reference parameter group, the required stress compensation amount is calculated to reduce or increase the air pressure to ensure a smoother conveying process of the screws:

[0073] ;

[0074] Among them, is the stress compensation coefficient, which is used to convert the stress difference into the air pressure compensation amount. The conveying cycle deviation and the stress compensation amount are input into the adaptive algorithm module, and through weighted calculation, a final air pressure correction value is obtained:

[0075] ;

[0076] Among them, and are the weight coefficients of the conveying cycle deviation and the stress compensation amount respectively, which are used to control the influence degrees of these two factors during the air pressure correction process. This correction value dynamically adjusts the air pressure in the feeding pipeline to ensure that the system can adaptively respond to various changes occurring during the feeding process. The air pressure correction value is calculated for the difference with the current real-time air pressure data to obtain the compensation amount :

[0077] ;

[0078] This difference reflects the deviation between the actual air pressure and the expected air pressure of the current feeding system. According to this deviation, an air pressure regulation instruction is generated through compensation operation to adjust the air pressure of the compressed air feeding pipeline:

[0079] ;

[0080] Through the air pressure regulation instruction, the air pressure in the feeding pipeline is adjusted in real time, so that the feeding process is always in an optimal state to ensure that each screw can be accurately and smoothly conveyed in place, guaranteeing the high efficiency and reliability of the entire locking process.

[0081] In an example, the air pressure regulation instruction is passed into the electric screwdriver driver, and a torque-rotation speed-temperature mapping function is constructed by combining the real-time displacement data collected by the angular displacement sensor, and the torque output parameters are calculated through the least squares method, including:

[0082] The air pressure regulation instruction is input into the electric screwdriver driver, and data segmentation is performed according to the preset pressure range to obtain the electric screwdriver control reference value;

[0083] The rotation state of the electric screwdriver chuck is detected in real time by the angular displacement sensor, the real-time displacement data is obtained through the signal conditioning circuit, and the real-time displacement data is subjected to timing analysis, and the real-time rotation speed value and acceleration value of the electric screwdriver are obtained through differential calculation;

[0084] The ambient temperature data collected by the temperature sensor is input into the compensation calculation unit, and the temperature compensation factor is obtained through correction operation;

[0085] Based on the electric screwdriver control reference value and the real-time rotation speed value, a speed response equation is constructed, and the temperature compensation factor is substituted into the equation to obtain the corrected rotation speed curve;

[0086] According to the corrected rotation speed curve and the acceleration value, a dynamic characteristic equation is established, and the torque-rotation speed-temperature mapping function is obtained through function conversion;

[0087] Perform the least squares method calculation on the torque - speed - temperature mapping function, obtain the initial torque value sequence through error iterative operation, and perform weighted average calculation on the initial torque value sequence to obtain the torque output parameter.

[0088] In this example, the generated air pressure regulation instruction is input into the electric screwdriver driver to precisely control the working state of the electric screwdriver during the locking process. The input air pressure regulation instruction represents the pressure change during the feeding process and is used to adjust the movement state of the electric screwdriver chuck under different load conditions. To achieve precise control of the air pressure, the air pressure data is segmented according to the preset pressure range of the system to enable more refined control of the electric screwdriver. The set air pressure range is to , and the control reference value of the electric screwdriver is obtained through segmentation , which is used to guide the operation of the electric screwdriver under different pressure conditions. The control reference value after segmentation is determined by the following formula:

[0089] ;

[0090] where represents the segmentation coefficient, whose range is between 0 and 1. By adjusting the segmentation coefficient , flexible control of different pressure intervals is achieved. To monitor the movement state of the electric screwdriver chuck, an angular displacement sensor is used to collect the rotation state of the electric screwdriver chuck in real time. The original displacement signal obtained by the angular displacement sensor is processed by the signal conditioning circuit to eliminate environmental noise and signal distortion, and accurate real - time displacement data is obtained. To analyze the movement state of the electric screwdriver, time - series analysis is performed on these displacement data, and the real - time rotational speed of the electric screwdriver is obtained through first - order difference calculation :

[0091] ;

[0092] Furthermore, by performing second - order difference calculation on the rotational speed, the real - time acceleration of the electric screwdriver is obtained

[0093] ;

[0094] where represents the instantaneous rotational speed of the chuck at time , and represents the instantaneous angular acceleration of the electric screwdriver. By calculating and monitoring these parameters, the changes in the speed and acceleration of the electric screwdriver during the locking process are grasped in real time. At the same time, the ambient temperature data is collected through a temperature sensor , it is input into the compensation calculation unit to eliminate the adverse effects of temperature on torque and rotational speed. Through the compensation function, a temperature compensation factor is calculated , which is used to correct the temperature deviation of the system. The calculation formula for the compensation factor is as follows:

[0095] ;

[0096] Among them, represents the temperature sensitivity coefficient, represents the reference temperature. When the ambient temperature deviates from the reference temperature, the compensation factor will be adjusted accordingly, so as to ensure that the torque and rotational speed outputs of the electric screwdriver can remain stable under different temperature conditions. Based on the control reference value of the electric screwdriver and the real-time rotational speed , a speed response equation is constructed. This equation is used to describe the relationship between the rotational speed of the electric screwdriver and the applied air pressure, and its form is:

[0097] ;

[0098] Among them, represents the moment of inertia of the electric screwdriver, represents the damping coefficient of the system, represents the air pressure loss caused by friction and other reasons. Substitute the temperature compensation factor into the speed response equation to obtain the corrected rotational speed curve, which reflects the dynamic response characteristics of the electric screwdriver under different temperature and pressure conditions. Based on the corrected rotational speed curve and the acceleration , a dynamic characteristic equation of the electric screwdriver is established. This equation describes the mechanical characteristics of the electric screwdriver during the locking process, and its specific form is:

[0099] ;

[0100] Among them, represents the torque output of the electric screwdriver at time . Through the joint analysis of the speed response equation and the dynamic characteristic equation, a mapping function including the relationship between torque, rotational speed and temperature is established, that is, the torque-rotational speed-temperature mapping function. This function is used to describe the torque output characteristics of the electric screwdriver under different rotational speed and temperature conditions. In order to optimize the torque output, the least squares method is used to calculate the torque-rotational speed-temperature mapping function. The least squares method is a data fitting technique. By fitting the actually collected data, the function form that best fits these data is found, and the initial torque sequence is obtained, denoted as The initial value sequence represents the initial torque output of the electric screwdriver under different working conditions. Through error iterative operations, these initial values are adjusted multiple times to continuously reduce the deviation between the calculated value and the actual value, thereby improving the fitting accuracy. The torque initial value sequence is weighted and averaged to obtain the final torque output parameter :

[0101] ;

[0102] where represents the weight of the th initial value. By assigning different weights to each initial value, different working conditions are tilted during the calculation to obtain a torque output that better conforms to the dynamic characteristics of the system

[0103] In one example, an adaptive compensation calculation is performed on the torque output parameter to obtain a torque compensation parameter, and the torque compensation parameter is integrated with the data of the locking angle sensor to generate a torque-locking angle characteristic curve, including

[0104] Sampling and analyzing the torque output parameter, calibrating the data based on the torque range of the electric screwdriver to obtain a torque reference sequence, and constructing a compensation calculation model based on the torque reference sequence. Substituting the no-load speed data into the compensation calculation model to obtain the torque compensation parameter

[0105] Dynamically compensating the torque accuracy according to the torque compensation parameter, and obtaining the compensated torque value through data correction

[0106] Sampling the data of the locking angle sensor collected by the locking angle sensor, obtaining the effective angle value through data screening, and performing time series matching on the compensated torque value and the effective angle value to obtain the torque-angle data pair through data alignment

[0107] Classifying the torque-angle data pairs according to the thread type, and obtaining the characteristic interval through data segmentation

[0108] Performing integral calculation on the data in the characteristic interval, obtaining the characteristic curve parameters through cumulative operation, and performing function fitting based on the characteristic curve parameters to generate the torque-locking angle characteristic curve through curve reconstruction

[0109] In this example, the torque output parameter of the electric screwdriver is sampled and analyzed to ensure that the torque generated during the operation of the electric screwdriver can accurately match the preset torque requirements. To better control the torque output, data calibration is performed based on the torque range of the electric screwdriver. The maximum and minimum torques of the electric screwdriver are denoted as and , respectively. These values are usually determined according to the design parameters and actual usage conditions of the electric screwdriver. By sampling the real-time torque output data of the electric screwdriver , a reference sequence of torque is obtained based on this data , and these reference data are used as the basis for subsequent compensation calculations. By comparing and calibrating the torque data collected in real time with a preset torque range, a torque reference sequence applicable to different working conditions is obtained:

[0110] ;

[0111] Substitute the no-load speed data into the constructed compensation calculation model to determine the compensation for the torque output caused by external factors (such as air pressure changes, friction, etc.). The no-load speed is usually the speed data recorded when the electric screwdriver motor runs idle without load, reflecting the speed response of the electric screwdriver under different air pressures and environmental conditions. The compensation calculation model adopts the following form:

[0112] ;

[0113] Where is the compensation coefficient, is the calibrated speed, indicating the speed of the electric screwdriver under standard load, is the torque value after compensation. Through this step, dynamic compensation is performed on the torque output of the electric screwdriver to make it more accurately match the set working conditions. Data correction is performed to ensure that the compensated torque value is effective in actual operation. Data correction is carried out by comparing with the torque data in the actual screw locking process. Suppose an actual torque data sequence is obtained through experiments. By comparing these actual data with the compensated torque value , the following error correction formula is used for data correction:

[0114] ;

[0115] Where is the correction error, which is calculated by the least squares method or other error correction methods. When performing data correction, considering the changes in different screws and assembly environments, multiple experiments are required to determine the appropriate error correction amount. The corrected torque value is used as the basis for subsequent analysis. Perform correlation analysis between the angle and the torque. During the screw locking process, the locking angle is a key parameter that directly affects the tightening quality of the screw. By using a locking angle sensor, the angle data during the locking process is collected in real time. These data are screened and cleaned to remove noise and invalid data, and the effective angle value is obtained. The screening process is carried out by setting a threshold and , exclude data that does not meet the actual operating conditions. For example, in some cases, the locking angle may show abnormal fluctuations, and the screening step ensures that only valid angle values are used for subsequent analysis. The compensated torque value and the effective locking angle value are matched in time series, and torque-angle data pairs are obtained through data alignment. The torque-angle data pairs are classified according to the thread type. The thread type of the screw has a great influence on the torque-angle relationship during the locking process, and classification is performed according to different thread types in data processing. Thread types include different pitches, thread angles, etc., which will all affect the mechanical properties during the locking process. Assume that the data is divided into categories, and the torque-angle data pairs are processed in segments according to different thread types to obtain the characteristic intervals under each category, where represents the classification index of different thread types. After obtaining the characteristic intervals of each type of data, the characteristic curve parameters within each interval are obtained through integral calculation. The integral calculation uses the method of accumulation, and the data within each characteristic interval is accumulated to obtain the characteristic parameters of this interval:

[0116] ;

[0117] where, represents the relationship function between torque and angle, which is set as a polynomial, exponential function or other forms of functions according to the actual situation. Based on the characteristic curve parameters , the data within the characteristic interval is fitted through function fitting to obtain a continuous torque-locking angle characteristic curve. Assume that the fitting function is , then the final torque-locking angle characteristic curve is obtained through the following formula:

[0118] ;

[0119] This characteristic curve can accurately describe the relationship between torque and locking angle under different thread types.

[0120] In an example, based on the torque-locking angle characteristic curve, the locking state is evaluated to obtain state evaluation indicators. The deviation analysis is performed between the state evaluation indicators and the set locking threshold, and the system correction matrix is obtained through multi-dimensional parameter collaborative calculation, including:

[0121] Perform data analysis on the torque-locking angle characteristic curve, extract key nodes according to the screw assembly state to obtain the locking state sequence, and perform online monitoring of the torque accuracy according to the locking state sequence. The torque state data is obtained through torque range analysis;

[0122] Input the torque state data into the evaluation unit, obtain the state evaluation index based on the fault alarm and counting function, monitor the three dimensions of material shortage, material jamming, and abnormal torque during the locking process, and obtain the abnormal identification matrix through threshold detection;

[0123] Fuse the state evaluation index and the abnormal identification matrix, obtain the comprehensive state score through weighted operation, compare the comprehensive state score with the set locking threshold, and obtain the deviation parameter group through difference calculation;

[0124] Construct a multi-dimensional state space based on the deviation parameter group, obtain the parameter correction amount through collaborative analysis, perform dimensionality reduction processing on the parameter correction amount according to the dynamic model, and obtain the system correction matrix through matrix transformation.

[0125] In this example, data analysis is performed on the torque-locking angle characteristic curve to extract the key locking nodes. These key nodes represent the significant change points in the relationship between torque and angle during the screw locking process, such as when the screw contacts the workpiece, when the screw starts to press in, when the predetermined tightening force is reached, etc. Assume that the obtained torque-locking angle data pairs are , where represents the torque value at the th moment, and is the corresponding locking angle value. By analyzing these data, these key nodes are extracted through numerical differentiation, trend detection, or inflection point identification methods. These key nodes reflect the locking state of the screw. After organizing these nodes, the locking state sequence is obtained, where each state corresponds to a specific stage of the locking process, such as "initial contact", "locking start", "torque approaching the set value", or "locking completed", etc. Based on the locking state sequence, on-line monitoring of the torque accuracy is performed. By analyzing the torque range, that is, by comparing the real-time torque data with the preset maximum torque and the minimum torque , the torque state data is obtained, which reflects whether the torque during the current screw assembly process is within the normal range. The calculation of the torque state data is expressed as:

[0126]

[0127] Among them, 1 indicates normal torque and 0 indicates abnormal torque. By monitoring the torque status data, it is possible to grasp in real time whether there are any deviations during the screwing process. The torque status data is input into the evaluation unit for status evaluation. Based on the real-time data, the quality of the entire screw screwing process is evaluated, and it is determined whether there are any faults or abnormalities. During this process, the fault alarm and counting functions are key parts. Based on these functions, status evaluation indicators in multiple dimensions are obtained. For example, during the screwing process, there are three typical abnormal situations: material shortage, material jamming, and abnormal torque. Material shortage refers to insufficient screw supply, resulting in the screw being unable to enter the chuck normally or unable to continue screwing; material jamming refers to the screw being blocked during transportation or screwing, resulting in the transportation or screwing process being blocked; and abnormal torque indicates that during the screwing process, the torque data exceeds the preset range, indicating a deviation in the screwing operation. According to the data in these three dimensions, a threshold detection method is used to detect abnormal situations in each dimension. Assume that the set thresholds are and For the angle, and For the torque, by comparing the real-time data with these thresholds, an abnormal identification matrix is constructed, and its value represents the abnormal situation in each current dimension. According to this matrix, the abnormal identification in each dimension at each moment is obtained. The status evaluation indicators are fused with the abnormal identification matrix. The abnormal information in each dimension is integrated, and a comprehensive status score is evaluated. Assume that the weight of each dimension is , corresponding to material shortage, material jamming, and abnormal torque respectively. The comprehensive status score is calculated by the following formula:

[0128] ;

[0129] where, is the abnormal identification at a specific time point, representing the status of material shortage, material jamming, and abnormal torque. If is greater than the set screwing threshold , it indicates that there are significant abnormalities in the system and further adjustment or alarm handling is required. If , it indicates that the current system status is normal. By calculating the difference between the comprehensive status score and the set screwing threshold, a deviation parameter group is obtained for further optimization and correction. The deviation parameter group represents the gap between the current system and the ideal state, and is calculated as:

[0130] ;

[0131] This deviation parameter group reflects the errors in the current screw locking process. Based on this, parameter correction is carried out to enable the system to reach the optimal working state again. To achieve this goal, a multi-dimensional state space is constructed based on the deviation parameter group, and the parameter correction amount is obtained through collaborative analysis. The construction of the multi-dimensional state space is usually achieved by integrating the deviations in each dimension to form a multi-dimensional vector. Each dimension in this vector represents the deviation in a specific aspect of the system, such as material shortage, material jamming, or abnormal torque. Through collaborative analysis, a correction amount vector is obtained, which indicates how the system should be adjusted in each dimension. Suppose the correction amount vector is , representing the correction amounts for material shortage, material jamming, and abnormal torque. By performing dimensionality reduction on the correction amount vector, it is simplified into a single correction amount, and a matrix transformation method is used to complete this operation. By performing dimensionality reduction on the correction amount vector, the final system correction matrix is obtained, which represents the correction amounts that the system needs to make on each control parameter. Through matrix transformation, the calculation formula for the correction matrix is:

[0132] ;

[0133] where is a weighting matrix representing the weights of each control parameter. By adjusting these control parameters, the performance of the system is restored to the ideal state.

[0134] In an example, the system correction matrix is input into the Lagrangian control model to update the dynamic weight coefficients, and an optimized control instruction set is output, including:

[0135] Substitute the system correction matrix into the motion equation of the electric screwdriver torque output system and the air pressure - rotational speed function relationship, and obtain the initial weight data through matrix operations;

[0136] Perform eigenvalue analysis on the initial weight data, optimize the weights in combination with the control reference parameter group to obtain the updated coefficient matrix, and perform weight iterative operations on the updated coefficient matrix to obtain the dynamic weight coefficients;

[0137] Perform online calibration on the dynamic weight coefficients according to the feeder rotational speed and pressure parameters to obtain the optimized control parameter group, and substitute the optimized control parameter group into the Lagrangian control model to obtain the control variable sequence through parameter mapping;

[0138] Perform parsing processing on the control variable sequence, and perform parameter matching in combination with the torque - locking angle characteristic curve to obtain the reference instruction set;

[0139] The reference instruction set is functionally divided according to electric screwdriver control and feeding control. Sub-control instructions are obtained through instruction parsing, and the sub-control instructions are dynamically combined. An optimized control instruction set is output through coordinated operations.

[0140] In this example, the system correction matrix is obtained through multi-dimensional analysis and collaborative calculation of the system state. This matrix contains all the control parameters that need to be adjusted, such as torque, rotational speed, air pressure, etc. Substituting this matrix into the motion equation of the electric screwdriver torque output system and the air pressure-rotational speed function relationship, initial weight data is obtained. Assume the motion equation of the electric screwdriver torque output system is:

[0141] ;

[0142] where, is the total force applied to the system, is the mass, a is the acceleration, is the damping matrix, is the velocity. The air pressure-rotational speed function relationship is expressed as:

[0143] ;

[0144] where, is the rotational speed, is the air pressure, and the function describes the relationship between air pressure and rotational speed. By substituting the system correction matrix into the above equations, initial control weight data is obtained, indicating the influence of each control parameter on the system's dynamic response. The initial weight data is represented as a vector , and each element represents the contribution of a parameter to the system performance. Perform eigenanalysis on the initial weight data and optimize the weights in combination with the control reference parameter group. The purpose of eigenanalysis is to discover the patterns and importance in the initial weight data, evaluate the contribution of each control parameter, and adjust the weights accordingly. Assume the control reference parameter group contains preset ideal control parameters, which can be used as references for optimization. The eigenanalysis method uses principal component analysis or other statistical analysis methods to project the initial weight data into the feature space of the reference parameter group to obtain the importance score of each control parameter. After eigenanalysis, the updated coefficient matrix represents the optimized control weights. The calculation process of the updated coefficient matrix is iteratively solved through methods such as the least squares method or the gradient descent algorithm, and the formula is expressed as:

[0145] ;

[0146] where, is the learning rate, is the loss function for the weights gradient. The loss function reflects the error between the current control system and the preset target. During the weight iteration process, by continuously updating the weight data, the dynamic weight coefficients are obtained, and these coefficients are used to describe the final influence of each control parameter (such as air pressure, rotational speed, etc.) in the system on the system performance. The dynamic weight coefficients are corrected online according to the feeder rotational speed and pressure parameters. The weight coefficients are dynamically adjusted through real-time data, enabling the system to adapt to changes in the external environment. For example, the rotational speed and pressure of the feeder are affected by system load or other factors, and are dynamically adjusted through real-time sampled data. Assume the feeder rotational speed is , and the air pressure is , and the online correction is performed through the following relational expression:

[0147] ;

[0148] where is the correction coefficient, represents the dynamically weighted coefficient after correction. Through this online correction method, the system can adapt to changes in air pressure and rotational speed in real time and achieve more precise control. Substitute the optimized control parameter group into the Lagrangian control model to generate the final control instruction set. The Lagrangian control model is a mathematical model that describes the motion of the system, and the control instructions are derived through the variational method. The control instruction set is the core for the system to achieve precise control, and the optimized control parameters are mapped to specific control variables through parameter mapping. Assume the control variable sequence is , and these variables represent the specific signals output by the system controller, such as the torque, rotational speed of the electric screwdriver, or the air pressure of the feeder. After parsing and processing the control variable sequence, parameter matching is performed in combination with the torque-locking angle characteristic curve. The torque-locking angle characteristic curve is a curve representing the relationship between torque and locking angle, obtained through experiments or theoretical calculations. During the parsing and processing, the control variables are matched with the characteristic curve to ensure that the system can reach the ideal torque value at the preset locking angle. Through this matching process, the reference instruction set is obtained, and this instruction set contains all the basic control commands that need to be executed. The reference instruction set will be functionally divided according to electric screwdriver control and feeder control. The complex control tasks are decomposed into multiple subtasks, and each subtask corresponds to a control link in the system. For example, electric screwdriver control is divided into torque control and rotational speed control, while feeder control includes air pressure and feeder rotational speed control. By parsing the reference instruction set, the sub-control instructions for each subtask are obtained By dynamically combining these sub-control instructions and performing coordinated operations, an optimized control instruction set is obtained. This instruction set will be transmitted to the electric screwdriver driver and the feeder control system to achieve precise screw locking control.

[0149] Referring to Figure 2 , this embodiment provides a screw feeding pressure adaptive adjustment device, including:

[0150] Acquisition module 1, used to extract screw specification parameters from the screw feeding device, and collect the electric screwdriver rotation speed signal and the initial torque signal to construct a Lagrangian control model, obtaining a control reference parameter group;

[0151] Calculation module 2, used to obtain the real-time air pressure data stream, and calculate the conveying cycle deviation and the screw impact stress value according to the control reference parameter group, and obtain the air pressure regulation instruction through difference compensation operation;

[0152] Construction module 3, used to transmit the air pressure regulation instruction to the electric screwdriver driver, jointly construct a torque-rotation speed-temperature mapping function with the real-time displacement data collected by the angular displacement sensor, and calculate the torque output parameter through the least square method;

[0153] Compensation module 4, used to perform adaptive compensation calculation on the torque output parameter, obtain the torque compensation parameter, and perform integral operation on the torque compensation parameter and the data of the locking angle sensor to generate a torque-locking angle characteristic curve;

[0154] Analysis module 5, used to evaluate the locking state based on the torque-locking angle characteristic curve, obtain the state evaluation index, perform deviation analysis on the state evaluation index and the set locking threshold, and obtain the system correction matrix through multi-dimensional parameter collaborative calculation;

[0155] Output module 6, used to input the system correction matrix into the Lagrangian control model, update the dynamic weight coefficient, and output the optimized control instruction set.

[0156] In this embodiment, for the specific implementation of each unit in the above device embodiment, please refer to that described in the above method embodiment, and details are not described herein again.

[0157] Referring to Figure 3 , this embodiment of the present invention also provides a computer device, which can be a server, and its internal structure can be as Figure 3As shown. The computer device includes a processor, a memory, a display screen, an input device, a network interface, and a database connected via a system bus. Among them, the processor of the computer design is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program, and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store the corresponding data in this embodiment. The network interface of the computer device is used to communicate with an external terminal via a network connection. The computer program, when executed by the processor, implements the above method.

[0158] Those skilled in the art can understand that Figure 3 the structure shown in is only a block diagram of a part of the structure related to the solution of the present invention, and does not constitute a limitation on the computer device to which the solution of the present invention is applied.

[0159] An embodiment of the present invention also provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the above method is implemented. It can be understood that the computer-readable storage medium in this embodiment can be a volatile readable storage medium or a non-volatile readable storage medium.

[0160] Those of ordinary skill in the art can understand that all or part of the processes in the methods of the above embodiments can be completed by instructing relevant hardware through a computer program. The computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above methods. Among them, any reference to a memory, storage, database, or other medium provided by the present invention and used in the embodiments can include non-volatile and / or volatile memories. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory can include random access memory (RAM) or an external cache memory. By way of illustration and not limitation, RAM is available in various forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (SSRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM, etc.

[0161] It should be noted that in this text, the terms "comprising", "including" or any other variants thereof are intended to cover non-exclusive inclusion, such that a process, apparatus, article or method comprising a series of elements not only includes those elements but also includes other elements not expressly listed, or further includes elements inherent to such process, apparatus, article or method. Without further limitation, an element defined by the phrase "comprising an..." does not exclude the presence of additional identical elements in the process, apparatus, article or method comprising such element.

[0162] The above are only the preferred embodiments of the present invention, and do not limit the patent scope of the present invention accordingly. Any equivalent structure or equivalent process transformation made by using the content of the specification and drawings of the present invention, or directly or indirectly applied in other related technical fields, shall be similarly included in the patent protection scope of the present invention.

Claims

1. A method for adaptively adjusting the feeding pressure of screws, characterized in that, Including the following steps: Extract screw specification parameters from the screw feeding device, collect the electric screwdriver rotation speed signal and the initial torque signal, and construct a Lagrangian control model to obtain a control reference parameter set; Obtain the real-time air pressure data stream, calculate the conveying cycle deviation and the screw impact stress value according to the control reference parameter set, and obtain the air pressure regulation instruction through differential compensation operation; Input the air pressure regulation instruction into the electric screwdriver driver, construct a torque-rotation speed-temperature mapping function in combination with the real-time displacement data collected by the angular displacement sensor, and calculate the torque output parameter by the least squares method; Perform adaptive compensation calculation on the torque output parameter to obtain a torque compensation parameter, and perform integral operation on the torque compensation parameter and the data of the locking angle sensor to generate a torque-locking angle characteristic curve; Evaluate the locking state based on the torque-locking angle characteristic curve to obtain a state evaluation index, perform deviation analysis on the state evaluation index and the set locking threshold, and obtain a system correction matrix through multi-dimensional parameter collaborative calculation; Input the system correction matrix into the Lagrangian control model, update the dynamic weight coefficient, and output an optimized control instruction set.

2. The screw feeding pressure adaptive adjustment method according to claim 1, wherein The step of extracting screw specification parameters from the screw feeding device, collecting the electric screwdriver rotation speed signal and the initial torque signal, and constructing a Lagrangian control model to obtain a control reference parameter set includes: Enter the screw specification parameters through the input interface of the liquid crystal screen control system, classify the data according to the preset specification range, and obtain the initial screw parameter matrix; Calculate the ratio of the total length L of the screw to the nut diameter D in the initial screw parameter matrix, screen out the effective parameter group with L / D greater than 1.

3. At the same time, digitally process the no-load rotation speed signal of the electric screwdriver collected by the encoder, and obtain the rotation speed digital quantity through signal sampling conversion; Perform differential operation on the displacement data collected by the angular displacement sensor to obtain the real-time rotation speed change rate, input the temperature data collected by the temperature sensor into the compensation function to obtain the initial torque coefficient, and perform temperature correction on the initial torque coefficient to obtain the corrected torque coefficient; Based on the effective parameter group, the rotation speed digital quantity, and the rotation speed change rate, establish the motion equation of the electric screwdriver torque output system, sample the rotation speed and pressure parameters of the feeder in the compressed air pipeline, and obtain the air pressure-rotation speed function relationship through data fitting; Substitute the motion equation of the electric screwdriver torque output system, the corrected torque coefficient, and the air pressure-rotation speed function relationship into the Lagrangian control model, construct a dynamic model, and obtain a control reference parameter set.

3. The screw feeding pressure adaptive adjustment method according to claim 2, characterized in that The step of obtaining the real-time air pressure data stream, calculating the conveying cycle deviation and the screw impact stress value according to the control reference parameter set, and obtaining the air pressure regulation instruction through differential compensation operation includes: Collect data through the distributed pressure sensor set in the compressed air supply pipeline, obtain the initial air pressure data through signal conversion, and input the initial air pressure data into the digital filter to eliminate environmental noise interference to obtain the real-time air pressure data stream; Perform a periodic analysis on the real-time air pressure data stream according to the feeder rotation speed parameter in the control reference parameter group to obtain the conveying time value of a single screw, and perform a periodic calculation based on the conveying time value of the single screw. Compare the calculation result with the preset standard period of 0.6 seconds to obtain the conveying period deviation; Detect the acting force when the screw reaches the electric screwdriver chuck through a high-precision torque sensor, obtain the screw impact stress value through stress conversion, and perform a threshold judgment on the screw impact stress value. Calculate the stress compensation amount based on the air pressure function in the control reference parameter group; Input the conveying period deviation and the stress compensation amount into the adaptive algorithm module, obtain the air pressure correction value through weighted calculation, and generate an air pressure regulation command through compensation operation according to the difference between the air pressure correction value and the current air pressure data.

4. The screw feeding pressure adaptive adjustment method according to claim 3, characterized in that The air pressure regulation command is transmitted to the electric screwdriver driver, and a torque-rotation speed-temperature mapping function is constructed by combining the real-time displacement data collected by the angular displacement sensor. The torque output parameter is calculated through the least squares method, including: Input the air pressure regulation command into the electric screwdriver driver, and perform data segmentation according to the preset pressure range to obtain the electric screwdriver control reference value; Real-time detect the rotation state of the electric screwdriver chuck through an angular displacement sensor, obtain the real-time displacement data through a signal conditioning circuit, and perform a timing analysis on the real-time displacement data. Obtain the real-time rotation speed value and acceleration value of the electric screwdriver through differential calculation; Input the ambient temperature data collected by the temperature sensor into the compensation calculation unit, and obtain the temperature compensation factor through correction operation; Construct a speed response equation based on the electric screwdriver control reference value and the real-time rotation speed value, and substitute the temperature compensation factor into the equation to obtain the corrected rotation speed curve; Establish a dynamic characteristic equation according to the corrected rotation speed curve and the acceleration value, and obtain the torque-rotation speed-temperature mapping function through function conversion; Perform the least squares method calculation on the torque-rotation speed-temperature mapping function, obtain the torque initial value sequence through error iterative operation, and perform weighted average calculation on the torque initial value sequence to obtain the torque output parameter.

5. The screw feeding pressure adaptive adjustment method according to claim 4, wherein Perform adaptive compensation calculation on the torque output parameter to obtain the torque compensation parameter, and perform integral operation on the torque compensation parameter and the data of the locking angle sensor to generate a torque-locking angle characteristic curve, including: Perform sampling analysis on the torque output parameter, perform data calibration based on the electric screwdriver torque range to obtain the torque reference sequence, and construct a compensation calculation model based on the torque reference sequence. Substitute the no-load rotation speed data into the compensation calculation model to obtain the torque compensation parameter; Dynamically compensate the torque accuracy according to the torque compensation parameter, and obtain the compensated torque value through data correction; Sample the data of the locking angle sensor collected by the locking angle sensor, obtain the effective angle value through data screening, and perform timing matching on the compensated torque value and the effective angle value. Obtain the torque-angle data pair through data alignment; Classify the torque-angle data pairs according to the thread type, and obtain the characteristic interval through data segmentation; Integrate the data within the characteristic interval, obtain the characteristic curve parameters through cumulative operations, and perform function fitting based on the characteristic curve parameters to generate a torque-locking angle characteristic curve through curve reconstruction.

6. The screw feeding pressure adaptive adjustment method according to claim 5, characterized in that, Based on the torque-locking angle characteristic curve, evaluate the locking state to obtain a state evaluation index, perform deviation analysis on the state evaluation index and the set locking threshold, and obtain a system correction matrix through multi-dimensional parameter collaborative calculation, including: Perform data analysis on the torque-locking angle characteristic curve, extract key nodes according to the screw assembly state to obtain a locking state sequence, and perform online monitoring of the torque accuracy according to the locking state sequence to obtain torque state data through torque range analysis; Input the torque state data into the evaluation unit, obtain a state evaluation index based on the fault alarm and counting function, monitor the three dimensions of material shortage, material jamming, and abnormal torque during the locking process, and obtain an abnormal identification matrix through threshold detection; Fuse the state evaluation index and the abnormal identification matrix, obtain a comprehensive state score through weighted operations, compare the comprehensive state score with the set locking threshold, and obtain a deviation parameter group through difference calculation; Construct a multi-dimensional state space based on the deviation parameter group, obtain a parameter correction amount through collaborative analysis, perform dimensionality reduction processing on the parameter correction amount according to the dynamic model, and obtain a system correction matrix through matrix transformation.

7. The screw feeding pressure adaptive adjustment method according to claim 6, characterized in that Input the system correction matrix into the Lagrangian control model, update the dynamic weight coefficient, and output an optimized control instruction set, including: Substitute the system correction matrix into the motion equation of the electric screwdriver torque output system and the air pressure-rotation speed function relationship, and obtain initial weight data through matrix operations; Perform characteristic analysis on the initial weight data, optimize the weights in combination with the control reference parameter group to obtain an updated coefficient matrix, and perform weight iteration operations on the updated coefficient matrix to obtain a dynamic weight coefficient; Perform online calibration on the dynamic weight coefficient according to the feeder rotation speed and pressure parameters to obtain an optimized control parameter group, substitute the optimized control parameter group into the Lagrangian control model, and obtain a control variable sequence through parameter mapping; Perform parsing processing on the control variable sequence, perform parameter matching in combination with the torque-locking angle characteristic curve to obtain a reference instruction set; Divide the reference instruction set according to the functions of electric screwdriver control and feeder control, obtain sub-control instructions through instruction parsing, and perform dynamic combination on the sub-control instructions to output an optimized control instruction set through coordinated operations.

8. A screw feeding pressure adaptive adjustment device, characterized in that, For implementing the steps of the method according to any one of claims 1 to 7, the device includes: An acquisition module for extracting screw specification parameters from the screw feeding device, and acquiring the electric screwdriver rotation speed signal and the initial torque signal to construct a Lagrangian control model to obtain a control reference parameter group; A calculation module for obtaining the real-time air pressure data stream, calculating the conveying cycle deviation and the screw impact stress value according to the control reference parameter group, and obtaining an air pressure regulation instruction through difference compensation operations; A building module for transmitting the air pressure regulation instruction to an electric screwdriver driver, constructing a torque-speed-temperature mapping function by combining the real-time displacement data collected by an angular displacement sensor, and calculating torque output parameters through the least squares method; A compensation module for performing adaptive compensation calculation on the torque output parameters to obtain torque compensation parameters, and performing integral operation on the torque compensation parameters and the data of a locking angle sensor to generate a torque-locking angle characteristic curve; An analysis module for evaluating the locking state based on the torque-locking angle characteristic curve to obtain a state evaluation index, performing deviation analysis on the state evaluation index and a set locking threshold, and obtaining a system correction matrix through multi-dimensional parameter collaborative calculation; An output module for inputting the system correction matrix into the Lagrangian control model, updating the dynamic weight coefficient, and outputting an optimized control instruction set.

9. A computer device, comprising a memory and a processor, wherein a computer program is stored in the memory, characterized in that, When the processor executes the computer program, the steps of the method according to any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 7 are implemented.

Images