Self-adaptive rivet pulling judgment method for electric riveting gun

Abstract

The invention relates to the field of electric tool control, and discloses a self-adaptive rivet pulling judgment method for an electric riveting gun, which is used for solving the problems of low judgment reliability and lack of self-adaptive capability caused by dependence on a fixed current threshold in the prior art. The method comprises the following steps: acquiring real-time current of a motor; calculating a real-time current difference value and comparing the real-time current difference value with a pop-rivet starting difference value threshold value to judge the start of operation; after operation is started, whether the current reaches a peak value and then drops to a pop-rivet successful drop threshold value is monitored to judge success, or whether the operation time exceeds the maximum timeout time is monitored to judge failure; and after the operation succeeds, extracting current characteristics of the operation for dynamically updating the threshold value. According to the method, the influence of null-pull current drift is avoided through current difference judgment, self-adaption to different working conditions is achieved by dynamically updating the threshold value, and the judgment reliability and universality are improved. In addition, failure reasons can be diagnosed, and operation efficiency is improved.

Description

Technical Field

[0001] This invention relates to the field of power tool control technology, and in particular to an adaptive rivet gun retrieval method. Background Technology

[0002] Electric rivet guns, as efficient joining tools, are widely used in many manufacturing fields. To ensure connection quality and achieve automated control, accurate judgment of the rivet operation process is necessary. In existing technologies, a common method is to judge the operating status by monitoring the magnitude of the motor drive current, as the current directly reflects the mechanical load on the motor.

[0003] However, this method of judgment based on the absolute value of the current has reliability issues in practical applications. This is because the no-load reference current of the motor is not constant; it is affected by various factors such as the heat generated by the equipment during long-term operation and changes in ambient temperature, causing it to drift. This drift can render the preset fixed threshold inapplicable, thus leading to misjudgments of the start or end of the operation.

[0004] Furthermore, when rivets of different specifications or materials are used, their tensile load characteristics change significantly, leading to a change in the current curve shape throughout the operation. Traditional fixed threshold methods lack the ability to adapt to these changes. To ensure accuracy, tedious manual resetting or calibration is often required, limiting the equipment's versatility and operational efficiency. Therefore, providing a rivet-pulling method that overcomes these shortcomings and achieves accurate and adaptive judgment is a pressing technical problem in this field. Summary of the Invention

[0005] The purpose of this invention is to provide an adaptive rivet gun rivet judgment method, which solves the problem that the existing judgment method based on a fixed current threshold is prone to low reliability and lacks adaptive capability due to the influence of no-load current drift and rivet specification changes.

[0006] To achieve the above objectives, the present invention provides the following technical solution: An adaptive rivet gun detection method for electric rivet guns includes the following steps: Obtain the real-time motor current of the electric rivet gun during operation; The real-time current difference is calculated based on the real-time motor current. When the real-time current difference is greater than the nail start difference threshold, the nail pulling operation is determined to start. After the rivet pulling operation is determined to have started, the real-time motor current is monitored. If the real-time motor current drops to less than or equal to the rivet pulling success drop threshold after reaching the peak current, the rivet pulling operation is determined to be successful. If the time elapsed since the start of the rivet pulling operation exceeds the maximum timeout time, the rivet pulling operation is determined to have failed. After determining that the rivet operation is successful, the rivet start difference threshold and the rivet successful descent threshold are updated based on the characteristics of the real-time motor current of the successful operation.

[0007] Preferably, the step of calculating the real-time current difference based on the real-time motor current specifically includes: The real-time motor current is continuously collected at a preset sampling frequency to obtain a current time series. The difference between the motor current value at the current sampling moment and the motor current value at the previous sampling moment is taken as the real-time current difference.

[0008] Preferably, the step of monitoring the real-time motor current after determining that the rivet operation has started further includes: After determining that the rivet operation has started, the maximum value of the real-time motor current is tracked in real time and recorded as the peak current.

[0009] Preferably, updating the rivet initiation difference threshold and the rivet successful descent threshold specifically includes: After determining that the rivet operation is successful, the trigger current difference at the start of the rivet operation and the peak current of the successful operation are extracted as feature information of the successful operation. The aforementioned feature information is stored in the successful job feature database; Based on the feature information of at least one successful operation stored in the successful operation feature database, the rivet start difference threshold and the rivet successful descent threshold are recalculated and updated.

[0010] Preferably, the update of the rivet start difference threshold is obtained by calculating the average value of the trigger current difference of at least one successful operation stored in the successful operation feature database, and then multiplying it by a preset safety factor.

[0011] Preferably, the update of the successful descent threshold of the pull pin is obtained by calculating the average value of the peak current of at least one successful operation stored in the successful operation feature database, and then multiplying it by a preset descent ratio coefficient.

[0012] Preferably, the method further includes: After determining that the rivet pulling operation has failed, the real-time motor current from the start of the rivet pulling operation to the time of the failure is analyzed to determine the rivet pulling failure mode.

[0013] Preferably, determining the rivet failure mode specifically includes: If the amplitude of the real-time motor current is higher than the product of the preset motor stall current limit and the preset stall coefficient during the maximum timeout period of the operation, and the duration exceeds the preset duration threshold, then the failure mode of the pull pin is determined to be a jammed or stalled mode.

[0014] Preferably, determining the rivet failure mode further includes: The historical average peak current is calculated based on the successful operation feature database. If the maximum value of the real-time motor current during the entire operation is less than the product of the historical average peak current and the preset judgment coefficient, then the failure mode of the pull nail is determined to be slippage or empty pull mode.

[0015] Preferably, after determining the rivet failure mode, the method further includes: Based on the determined pin failure mode, output the corresponding diagnostic information.

[0016] In summary, the present invention has at least one of the following beneficial technical effects: 1. This invention determines the start of rivet pulling by calculating the real-time current difference, rather than relying on the static absolute value of the current. This utilizes the physical characteristic that the instantaneous current surge upon load application makes the judgment independent of the magnitude of the no-load reference current. Therefore, no-load current drift caused by equipment heating, ambient temperature changes, or differences in rivet material between different batches will not interfere with the accuracy of the start-up determination, thereby improving the reliability and environmental adaptability of this invention under complex working conditions.

[0017] 2. This invention extracts the current characteristics of each successful operation and uses them to update the judgment threshold, forming a dynamic self-calibrating closed-loop system. This allows the judgment parameters to automatically adapt to changes in rivet specifications and the long-term operating conditions of the equipment, always maintaining the judgment benchmark within an optimal range. This eliminates the need for manual intervention and recalibration, ensuring high-precision judgment while significantly enhancing the long-term stability of the equipment and its adaptability to different operating objects.

[0018] 3. This invention, after determining that a task has failed, can further analyze the current curve during the failure process to identify the specific failure mode. By transforming the abstract timeout failure into specific and clear diagnostic information, it effectively guides operators to quickly locate and resolve the problem, thereby significantly shortening troubleshooting time and equipment downtime, while improving overall production efficiency. Attached Figure Description

[0019] Figure 1This is a schematic diagram of the method flow of the present invention. Detailed Implementation

[0020] The following is in conjunction with the appendix Figure 1 The present invention will be further described in detail below.

[0021] Before performing the core operation judgment, the method first performs system initialization in step S100. This step provides the initial parameter basis and necessary data structure for subsequent pin-pulling operation judgment, adaptive calibration, and intelligent diagnosis.

[0022] This step S100 may specifically include: S101, the controller loads a set of decision parameters from a non-volatile memory. This non-volatile memory can be an electrically erasable programmable read-only memory (EEPROM) or flash memory. This set of decision parameters provides an initial reference for subsequent job decisions and may include: Pin start-up difference threshold: This parameter is used in step S200 to compare with the real-time calculated current difference to identify whether the pin pulling operation has transitioned from an unloaded state to a loaded state. Its initial value can be an empirically set engineering value that can distinguish between unloaded current fluctuations and actual load start-up.

[0023] Successful pull-down threshold: This parameter is used in step S300 to determine whether the current has sufficiently decreased after reaching its peak value to confirm that the nail rod has been successfully pulled off.

[0024] Maximum job timeout: This parameter is used in step S300 as the upper limit of the time for judging job failure, so as to avoid the equipment running for a long time under abnormal operating conditions.

[0025] Motor stall current limit: This parameter is a physical characteristic parameter of the motor, representing the maximum current when the motor is stalled. It is used in subsequent failure mode diagnosis as a basis for determining whether the motor is stuck.

[0026] The above parameters can be loaded either the default values ​​preset at the factory or the latest values ​​saved after being updated by the dynamic threshold self-calibration process in the previous work cycle.

[0027] S102, the controller establishes a data structure in its internal random access memory (RAM) to store the characteristic information of successful nail pulling operations, namely the successful operation characteristic database. This database provides data support for subsequent dynamic threshold self-calibration and failure mode diagnosis.

[0028] The database can be implemented as a first-in, first-out (FIFO) queue with a preset capacity N, for example, N can be set to 10.

[0029] Each record stored in the database corresponds to a successful rivet operation, and each record contains a set of feature pairs that characterize the core shape of the current curve for that successful operation. In this embodiment, the feature pair is (ΔI trigger,i ,I peak,i ).

[0030] ΔI trigger,i This represents the current difference at the time of triggering the start-up judgment during the i-th successful operation.

[0031] I peak,i This represents the peak current monitored during the i-th successful operation.

[0032] Through initialization steps S101 and S102, the controller not only obtains all the parameters required to perform the initial judgment, but also prepares the foundation for data recording and adaptive learning in subsequent operations.

[0033] In step S200, the controller performs a rivet start determination. This step is one of the core innovations of this invention, as it achieves accurate identification of the rivet start moment by analyzing the dynamic rate of change of current, rather than its static absolute value. This step may further include: S201, upon receiving the pin-pulling command, the controller continuously samples the motor current at a preset sampling frequency, thereby obtaining a series of discrete current time series data I(t). The setting of this sampling frequency can be selected by those skilled in the art based on the processing power of the microcontroller used and the requirements for the immediacy of signal response; this is well-known technology in the field and will not be elaborated upon here.

[0034] S202, to accurately identify the actual current rise caused by the load from the background current fluctuations, the controller calculates the difference between the current value I(t) at each sampling time t and the current value I(t-Δt) at the previous sampling time, obtaining the real-time current difference ΔI(t). The calculation method is as follows: ΔI(t) = I(t) - I(t - Δt); In the formula, ΔI(t) is the current difference calculated at time t; I(t) is the motor current value collected in real time at time t; I(t-Δt) is the motor current value collected at the previous sampling time; and Δt is the time interval for current sampling.

[0035] The physical meaning of the current difference ΔI(t) lies in characterizing the rate of change of the current. The initiation of a rivet pulling operation physically involves the rivet gun's claws gripping the rivet shank and applying tension—a sudden, abrupt change from no-load to load. This abrupt change is directly reflected in the motor current, resulting in a rapid and significant jump. Therefore, by calculating the current difference, this method can sensitively capture this jump event.

[0036] Traditional methods rely on a fixed absolute current threshold for judgment, which is susceptible to drift in the no-load reference current. This drift can be caused by factors such as changes in ambient temperature, motor heating due to prolonged continuous operation, and differences in the material properties of rivets from different manufacturers or batches. These factors all lead to variations in the no-load current when the tool is not pulling the rivet. This new method uses current difference for judgment. Since the difference calculation only focuses on the relative change in current over a short period and is independent of the absolute reference current value, it effectively avoids the interference of no-load current drift caused by the aforementioned factors on the start-up judgment.

[0037] S203, the controller compares the real-time current difference ΔI(t) calculated in S202 with the current pin start difference threshold Th, which was loaded in step S100 or updated in step S400. start Perform real-time comparisons.

[0038] If ΔI(t) < Th start The controller determines that the tool is still in an idle or idling state and continues to execute S201 to S203 in a loop.

[0039] If ΔI(t) > Th start The controller then determines that the rivet operation has officially begun. At this point, the controller immediately records the current timestamp as the operation start time T. start Then, it proceeds to the operation status monitoring stage in step S300.

[0040] After determining the start of the rivet pulling operation in step S200, the method immediately proceeds to the operation status monitoring stage in step S300. The principle of this monitoring stage is based on the deterministic mapping relationship between the physical process of rivet pulling and the curve shape of the motor drive current. During the rivet pulling operation, the rivet shank undergoes tension, necking, and eventual breakage, causing the motor load to experience a change from increase to peak value and then instantaneous unloading. This load change is directly reflected in the motor current exhibiting a characteristic pattern of first rapidly rising to a peak value and then rapidly decreasing. Therefore, the controller determines whether the operation has been successfully completed by monitoring whether the current curve fully exhibits the peak-to-decline characteristic.

[0041] Specifically, at the start time T of the operation startThen, the controller continues to execute two parallel judgment tasks: success condition judgment and failure condition judgment.

[0042] The controller will track the success criteria in real time. start The controller calculates the maximum current value since then and temporarily stores it as the peak current. After the current surpasses the peak and begins to decrease, the controller compares the real-time current value I(t) with the current threshold value Th, which is either loaded in step S100 or updated in step S400, to determine the successful decrease of the pull stud. success Compare them.

[0043] When I(t) ≤ Th success When the condition is met for the first time, the rivet operation is considered successful, and the post-success processing flow of step S400 is triggered.

[0044] The judgment of failure conditions is essentially a timeout protection mechanism. During the rivet pulling operation, if abnormal conditions occur such as rivet jamming, rivet head slippage, or a serious mismatch in rivet specifications, the current curve will not show a typical successful pattern, and the operation cannot be completed within the expected time. Therefore, the controller will synchronously calculate the time from the start time T of the operation. start The elapsed time, i.e., the time t of the task. elapsed The time t already used elapsed The calculation method is t elapsed =tT start If the time t already elapsed... elapsed The maximum timeout period T set in step S100 has been exceeded. max If the above success conditions are still not met, the controller determines that the rivet operation has failed and triggers the failure post-processing procedure in step S400.

[0045] By monitoring the success and failure conditions in parallel, it is ensured that each rivet operation has a clear termination state, avoiding infinite waiting or uncertain state of the system under abnormal operating conditions.

[0046] After the rivet operation is determined to be successful in step S300, the controller executes the post-success processing procedure in step S400. This procedure uses dynamic threshold self-calibration to adaptively adjust the key thresholds used in the core judgment stage according to the actual operation situation, thereby continuously optimizing the accuracy of the judgment.

[0047] The post-success processing procedure may specifically include: S411, the controller extracts features from the current curve data of the recently completed successful operation. The extracted features are key parameters that characterize the core characteristics of the operation. In this embodiment, this step extracts the following two features: The current difference at the start-up: This value is the value obtained when ΔI(t) > Th for the first time in step S203. start The current difference ΔI(t) under the given conditions.

[0048] Peak current of this operation: This value is calculated in step S300 from the start time T. start The maximum value of the motor current monitored between the time of successful operation determination and the time of successful operation determination.

[0049] S412, the controller will extract the feature pairs (ΔI) from S411. trigger,new ,I peak,new As a new record, it is stored in the successful job feature database D established in step S102. success The database employs a First-In-First-Out (FIFO) queue management method. If the database has reached its preset capacity N before storing a new record, the controller first removes the oldest record and then stores the new record. This mechanism ensures that the database always stores the characteristics of the N most recent successful jobs.

[0050] S413, the controller is based on database D success The current k records in the database are used to determine the threshold difference Th for starting the rivet. start And the rivet successfully lowered the threshold Th success Perform a recalculation and update.

[0051] For the tack start difference threshold Th start The update is calculated as follows: In the formula, Th start The updated trigger threshold is defined as follows: α is a preset safety factor less than 1, such as 0.8, which provides a margin for the new threshold based on the recent average trigger strength to ensure the sensitivity of the trigger detection; k is the number of records currently stored in the database; ΔI trigger,i This represents the current difference at the trigger start of the i-th record in the database.

[0052] For the rivet to successfully decrease the threshold Th success The update process can be divided into two steps. First, the average peak current of historical successful operations is calculated: In the formula, I peak,avg I represents the average peak current of k successful jobs in the database. peak,i Let be the peak current of the i-th record in the database.

[0053] Subsequently, Th is updated based on this average peak current. success: Th success =β·I peak,avg ; In the formula, Th success The updated pull stud success threshold is defined by β, which is a preset drop ratio coefficient less than 1, such as 0.6, representing the percentage of current drop from the peak value required to successfully break the stud. peak,avg Let be the average peak current of k successful jobs in the database.

[0054] S414, the controller will use the new Th calculated in S413. start and Th success The updated threshold values ​​are written to non-volatile memory for storage. In this way, these two updated thresholds can be used not only for the next nailing operation, but also as initial values ​​loaded after a power outage and restart, thus solidifying the system's adaptive learning results.

[0055] If a failure condition is triggered in step S300 because the elapsed time of the task exceeds the maximum timeout period, the controller executes the post-failure processing procedure in step S400. This procedure analyzes and diagnoses the current curve during the failure process, categorizes the cause of the failure into a preset pattern, and provides targeted feedback to the operator. This procedure is one of the supplementary innovations of this invention.

[0056] The failure handling process can specifically include: S421, the controller first retrieves and stores the complete current curve data during this failed operation. This data is from the start time T of the operation. start The time series of motor current collected between the timeout point and the timeout occurrence is denoted as I. failed (t).

[0057] S422, To perform diagnostic analysis, the controller needs to acquire a diagnostic baseline. This baseline may include: Average peak current of historical successful operations: This value is obtained by analyzing the characteristics of the successful operation database D. success The value is calculated from the data stored in the system, and the calculation method has been described in the post-success processing flow. This value represents the load level that should be achieved during normal operation.

[0058] Motor stall current limit: This value is applied in step S100 and represents the physical limit load capacity of the motor.

[0059] S423, the controller uses the failure current curve I obtained in S421. failed The morphological characteristics of (t) are analyzed and compared with the diagnostic benchmarks obtained in S422, thereby classifying the cause of failure into a preset failure mode. In this embodiment, the classification logic may include: Stuck or blocked mode: I failed (t) If the amplitude rises rapidly after the start of the operation and remains at a high level for a period of time before the timeout, and this level is close to or reaches the motor stall current limit, then the failure is classified as this mode. This mode usually corresponds to the situation where the internal mechanical structure of the gun head is stuck or the rivet is jammed by a foreign object.

[0060] Slip-off or pull-out mode: I failed (t) If, after triggering the start-up judgment, its overall amplitude remains at a low level throughout the entire operation time, and its maximum peak value is much smaller than the average peak current of historical successful operations, then this failure is classified as this mode. This mode usually corresponds to the situation where the gun head claw fails to effectively grasp the nail rod and slips out, or the gun head is pulled to the bottom without a rivet inside.

[0061] Rivet specification or material mismatch mode: I failed The curve (t) shows a peak, but this peak is significantly lower than the average peak current of historical successful operations. Furthermore, the current does not decrease rapidly after this peak to meet the success criteria until a timeout occurs. In this case, the failure is classified as this mode. This mode typically corresponds to the use of rivets with too small a diameter or too soft a material, causing abnormal deformation or breakage before reaching the normal breaking force. However, the resulting load reduction is insufficient to be considered a success.

[0062] S424: After S423 completes the classification of failure modes, the controller generates and outputs differentiated diagnostic information based on the specific classification results. This output can be one or more of the following combinations: status indication via indicator lights of different colors or flashing patterns; different beeping sounds via a buzzer; or direct display of specific text prompts on the device's display screen.

[0063] For example, for jammed or stalled modes, the system may prompt the operator to check for nozzle jamming; for slippage or free-pulling modes, it may prompt the operator to check if the rivet is correctly positioned; for modes with incorrect specifications or materials, it may prompt the operator to confirm the rivet specifications. This kind of clearly targeted diagnostic information can effectively guide operators in troubleshooting.

[0064] After completing the success or failure post-processing procedure in step S400, the method proceeds to the reset and standby stage in step S500. The purpose of this stage is to restore the mechanical components and software state of the electric rivet gun to their initial state, preparing for the next rivet operation.

[0065] This step S500 may specifically include: In step S501, the controller sends a command to the motor drive module to control the motor to perform a return-to-position action. This return-to-position action is typically a motor reversal. Its purpose is to drive the internal pulling mechanism of the rivet gun, such as the claw and pull rod, back to its initial position where it can receive new rivets. If the operation is successful, this action also coordinates the ejection of the broken rivet rod. The specific control implementation for motor reversal can be achieved by those skilled in the art using an H-bridge circuit or a corresponding integrated motor drive chip, which is well-known in the field and will not be elaborated upon here.

[0066] In step S502, during the execution or completion of the motor homing action, the controller clears or resets a series of state variables and temporary data used in the current job cycle. The variables being reset may include: a counter recording the elapsed job time, the peak current value recorded in the current job, and various status flags used to mark the job progress. This step ensures that the system's internal logic state is a defined initial state at the start of the next job, thereby preventing residual data from the previous job from interfering with the next judgment process.

[0067] After completing S501 and S502, a single complete work cycle of this method is concluded. The system then enters standby mode, continuously monitoring external pin pulling commands in preparation for starting the next work cycle.

Claims

1. A method for self-adaptive pull determination of an electric rivet gun, characterized in that, The method comprises the following steps: obtaining the real-time motor current of the motor of the electric rivet gun during operation; calculating the real-time current difference value based on the real-time motor current, and determining that the rivet operation starts when the real-time current difference value is greater than the rivet starting difference threshold value; monitoring the real-time motor current after determining that the rivet operation starts, and determining that the current rivet operation is successful when the real-time motor current drops to be less than or equal to the rivet successful drop threshold value after reaching the peak current value; updating the rivet starting difference threshold value and the rivet successful drop threshold value based on the characteristics of the real-time motor current of the successful operation after determining that the current rivet operation is successful.

2. The self-adaptive pull determination method for the electric rivet gun according to claim 1, characterized in that, The method further comprises: analyzing the real-time motor current from the start of the rivet operation to the moment when it is determined that the operation fails to determine the rivet failure mode after determining that the current rivet operation fails. The method further comprises:

3. The self-adaptive pull determination method for an electric rivet gun according to claim 1, characterized in that, calculating the historical average peak current value based on the successful operation characteristic database, and determining that the rivet failure mode is the slip or empty rivet mode when the maximum value of the real-time motor current in the entire operation process is less than the product of the historical average peak current value and a preset judgment coefficient. ​ 4. The self-adaptive pull determination method for a rivet gun according to claim 1, wherein, ​ ​ ​ ​ 5. The self-adaptive pull determination method for an electric rivet gun according to claim 4, wherein ​ 6. The self-adaptive pull determination method of the electric rivet gun according to claim 4, characterized in that, ​ 7. The self-adaptive pull determination method for a rivet gun according to claim 1, wherein, ​ ​ 8. The self-adaptive pull determination method of the electric rivet gun according to claim 7, characterized in that, ​ ​ 9. The self-adaptive pull determination method of the electric rivet gun according to claim 7, characterized in that, ​ ​ ​ 10. The self-adaptive pull determination method of the electric rivet gun according to claim 7, wherein, The determining the failure mode of the drawbar further includes: According to the determined failure mode of the drawbar, corresponding diagnostic information is output.

Images