CN-121995732-A - Self-adaptive control method and system for spin riveting gap of rotary fastener

CN121995732ACN 121995732 ACN121995732 ACN 121995732ACN-121995732-A

Abstract

The invention belongs to the technical field of spin riveting control, and discloses a spin riveting gap self-adaptive control method and system for a rotary fastener, wherein multi-dimensional key parameters of the fastener are comprehensively captured through a collection calibration stage, three core models of spin riveting gap dynamic prediction, gap rebound prediction and interference factor association are constructed by combining cold rolling physical characteristics of spin riveting processing, and real-time linkage of the parameters, the gap and the interference factor is realized through a self-adaptive adjustment algorithm; aiming at various dynamic interferences such as material parameter difference, tool wear, temperature fluctuation and the like, the method can accurately identify and execute targeted adjustment, avoid the problem of overlarge or overlarge gap caused by fixed parameters, improve the bonding strength, coaxiality and fatigue resistance of the rotary fastener, and accurately capture the dynamic change of the gap in the spin riveting process by adopting a displacement sensor, a torque sensor and a laser gap sensor cooperative detection system and performing double check of indirect detection and direct detection.

Inventors

ZHU SHILAI
Cao Huatong
ZHANG MIN

Assignees

南京方州金属制品有限公司

Dates

Publication Date: 20260508
Application Date: 20260206

Claims (9)

1. The self-adaptive control method for the spin riveting gap of the rotary fastener is characterized by comprising the following specific steps of: The method comprises the steps of collecting and calibrating a rotary fastener to be spin-riveted, sending the rotary fastener to be spin-riveted to a pretreatment station, synchronously collecting multidimensional key parameters of the fastener through detection equipment and completing spin-riveting positioning reference calibration; The model construction stage comprises the steps of constructing a rotary riveting gap dynamic prediction model based on collected basic data and combined with cold rolling physical characteristics of rotary riveting processing of a rotary fastener to realize the advance judgment of the gap variation, constructing a gap rebound prediction sub-model to pre-judge the rebound quantity and optimize the precision, constructing an interference factor correlation sub-model, and determining the corresponding relation between various interference factors and the gap interference quantity; The initial matching stage comprises the steps of combining database matching results, automatically matching spin riveting initial parameters through various model linkage fastener actual parameters and load grades, and conforming to a triple adaptation principle; The detection and identification stage comprises the steps of starting spin riveting operation and synchronously starting a multi-sensor collaborative detection system to realize multiple verification and filtration of gaps and interference, and enabling a linkage interference factor correlation sub-model to identify main interference factors in real time and output related information; The parameter adjusting stage comprises the steps of executing real-time self-adaptive adjustment of spin riveting parameters based on real-time gap data and interference information and combining various prediction models, comparing the real-time gap with a reference gap to calculate an offset value, and adopting a self-adaptive adjustment algorithm to adjust the spin riveting parameters; The rebound checking stage comprises the steps of carrying out rebound compensation on the gap by combining the rebound quantity pre-judged by the gap rebound prediction sub-model and the elastic recovery characteristic of the fastener material when the real-time gap is close to a reference value, finishing double judgment of gap qualification by detection equipment after unloading, returning to the parameter adjusting stage for readjustment if the gap qualification is unqualified, and recording the whole-flow data to a database if the gap qualification is qualified; And in the learning optimization stage, a reinforcement learning algorithm is adopted to analyze association rules among multidimensional parameters, spin riveting parameters, interference factors, gap control effects and load suitability of the fastener, optimize various models and database parameters, construct a spin riveting parameter optimization library, and add a tool wear early warning function.
2. The self-adaptive control method for the spin riveting gap of the rotary fastener according to claim 1 is characterized in that in the acquisition and calibration stage, special key parameters of the fastener and the space three-dimensional coordinates of a riveting reference surface are acquired through a multi-dimensional parameter synchronous acquisition system, wherein the special key parameters comprise the pore diameter form and position tolerance of the fastener, the wall thickness deviation value, the roughness of the riveting surface, the hardness of materials and the fluctuation value of elastic modulus, the association database associates the multi-dimensional parameters of the fastener, the reference gap and the load level, and an automatic recognition function of the type of the fastener is embedded, so that the integrated adaptation of the pre-detection, the automatic calibration of the reference gap and the load level matching of the fastener to be spin riveted is realized.
3. The spin riveting gap self-adaptive control method of the rotary fastener according to claim 2 is characterized in that in the model construction stage, a spin riveting gap dynamic prediction model takes fastener material hardness, elastic modulus fluctuation value and wall thickness deviation value as core input variables, a reference gap as target variables, spin riveting process parameter association coefficients and load grade weights are introduced, the spin riveting process parameter association coefficients and load grade weights are trained and optimized through an adaptation algorithm to improve convergence speed and prediction accuracy, a gap rebound predictor model is combined with material elasticity recovery characteristics and historical rebound data to predict rebound quantity and is incorporated into model correction term optimization accuracy, and a disturbance factor association submodel is used for determining the corresponding relation among tool wear, temperature fluctuation, material parameter deviation and gap disturbance quantity.
4. The self-adaptive control method for the spin riveting gap of the rotary fastener according to claim 3, wherein in the initial matching stage, spin riveting initial parameters comprise spin riveting head rotating speed, feeding speed, spin riveting pressure, spin riveting angle and pressure maintaining initial time, the triple adaptation principles comprise differential adaptation, load adaptation and interference pre-judgment adaptation, alignment of the spin riveting head and a riveting reference surface and sensor zero point calibration are completed when equipment is initialized, an initial parameter pre-verification step is added, parameter suitability is verified through small-range test riveting, and the deviation exceeds an allowable range and returns to the initial matching stage to re-match parameters.
5. The self-adaptive control method for the spin-rivet gap of the rotary fastener according to claim 4 is characterized in that in the detection and identification stage, multiple sensors cooperate to realize double check of indirect detection and direct detection, a linked interference factor correlation submodel realizes interference detection to form triple check, a displacement sensor indirectly reflects gap variation, a torque sensor supplements detection gap state and assists in judging interference, a laser gap sensor directly detects real-time gap value, a filtering algorithm is adopted to filter detection noise, a linked interference factor correlation submodel identifies main interference factors, and interference signal intensity, interference type and influence are output and fed back to a parameter adjustment link.
6. The self-adaptive control method for the spin riveting gap of the rotary fastener according to claim 5 is characterized in that a PID self-adaptive adjustment algorithm is adopted in the parameter adjustment stage to realize closed-loop control of gap detection, deviation calculation, interference identification and parameter adjustment, spin riveting pressure is increased, feeding speed is reduced and relevant process parameters are finely adjusted to reduce the gap when the real-time gap is larger than a reference gap, spin riveting pressure is reduced, feeding speed is increased and spin riveting angle is reduced when the real-time gap is smaller than the reference gap, targeted adjustment is carried out for different interference factors, and adjustment intervals adapt to structural characteristics and load requirements of the fastener.
7. The self-adaptive control method for the spin riveting gap of the rotary fastener according to claim 6, wherein in the rebound checking stage, when the real-time gap is close to a reference value, the spin riveting equipment reduces the pressure to enter an unloading stage, rebound compensation is realized by adjusting the dwell time and the unloading speed, the real-time gap is adjusted to a preset range in advance, the qualification judgment is combined with a final gap detection value and a torque detection value, the linkage rebound quantity and the interference factor are subjected to secondary fine adjustment when the final gap detection value and the torque detection value are failed, and the full-flow spin riveting data is recorded after the final gap detection value and the torque detection value are qualified.
8. The spin-rivet gap self-adaptive control method of the rotary fastener according to claim 7, wherein in the learning optimization stage, full-flow spin-rivet data comprise fastener multidimensional parameters, load levels, initial spin-rivet parameters, real-time adjustment parameters, gap deviation values, rebound amounts, interference factor types and influence amounts, detection results, tool wear data and environmental temperature fluctuation data, matching relations between model parameters and databases are optimized through reinforcement learning algorithm, and the spin-rivet parameter optimization library comprises special parameter adaptation schemes of fasteners with different types and different load levels.
9. A rotary fastener standoff adaptive control system based on the method of claim 8, comprising: the data acquisition calibration module is used for carrying multiple types of detection equipment and identification modules, completing the multi-dimensional parameter acquisition of the fastener, the positioning calibration of the riveting reference surface and the calibration of the reference spin riveting gap, establishing a correlation database and inputting tooling and environment basic data; the model prediction and initial matching module is used for running a spin riveting gap dynamic prediction model, a gap rebound predictor model and an interference factor correlation submodel, and the linkage database is used for completing spin riveting initial parameter adaptation, so as to realize spin riveting head alignment, equipment and sensor initialization and initial parameter pre-verification; the real-time detection and interference identification module is used for realizing double detection of gaps by means of a multi-sensor cooperative detection system, filtering noise by adopting a filtering algorithm, identifying interference factors in real time and outputting relevant information to be fed back to a parameter adjustment link; The parameter adjustment and rebound checking module adopts a PID self-adaptive adjustment algorithm to link detection data and a prediction model so as to realize real-time fine adjustment of spin riveting parameters; And the self-learning optimization module analyzes the whole flow data by adopting a reinforcement learning algorithm, optimizes the matching relation between each prediction model parameter and the database, constructs a spin riveting parameter optimization library and realizes tool wear early warning.

Description

Self-adaptive control method and system for spin riveting gap of rotary fastener Technical Field The invention belongs to the technical field of spin riveting control, and particularly relates to a spin riveting gap self-adaptive control method and system for a rotary fastener. Background The rotary fastener is widely applied to the fields of construction, machinery, aerospace and the like, and the spin riveting gap is used as a core control parameter for spin riveting processing, so that the joint strength, coaxiality and fatigue resistance after riveting are directly affected. At present, conventional metal spin riveting assembly modes are mostly adopted in spin riveting processing of the rotary fastener, but the following technical problems exist in the existing spin riveting gap control method: The existing spin riveting gap control mostly adopts a mode of presetting fixed technological parameters, parameters such as the rotational speed, pressure, feeding amount and the like of a spin riveting head are usually set in advance according to the type of a fastener, the parameters are kept unchanged in the whole processing process, and the processing errors of the spin fastener, physical parameter differences of materials and the influence of dynamic interference (such as riveting temperature change and tool abrasion) in the spin riveting process on the gap are not fully considered. Partial improvement scheme tries to carry out parameter feedback adjustment through single torque detection, but has the problems of overlong transmission chain and larger detection error, and the clearance rebound amount after spin riveting is not prejudged and compensated, and finally the situation that the clearance after spin riveting is too large or too small is caused, wherein the clearance is too large to cause loosening and falling of a fastener, the deformation and cracking of the fastener can be caused when the clearance is too small, and spin riveting equipment can be damaged when serious. The existing control method has the defects that the detection means are single, the parameter feedback adjustment is realized by mostly relying on single torque detection, the detection error is large, the dynamic change of the gap in the spin riveting process is difficult to accurately capture, the gap control precision is affected, and finally, the product percent of pass is lower. Disclosure of Invention The invention aims to provide a self-adaptive control method and a self-adaptive control system for a spin riveting gap of a rotary fastener, so as to solve the problems in the background technology. In order to achieve the purpose, the invention provides the following technical scheme that the spin-riveting clearance self-adaptive control method of the rotary fastener comprises the following specific steps: Preferably, the acquisition and calibration stage sends the rotary fastener to be riveted to a pretreatment station, a multi-dimensional parameter synchronous acquisition system is started, and special key parameters of the fastener, including pore diameter form and position tolerance, wall thickness deviation value, riveting surface roughness, material hardness and elastic modulus fluctuation value of the fastener, are synchronously acquired through a laser ranging sensor, a hardness detector, a visual identification module and an elastic modulus tester, and meanwhile, the spatial three-dimensional coordinates of a riveting reference surface of the fastener are acquired through a visual positioning algorithm, so that the calibration of the riveting positioning reference is completed; The data synchronous acquisition mechanism adopts a hardware triggering synchronous mode, the acquisition frequency of each detection device is uniformly set to 200Hz through uniform clock signals, the acquisition time difference of different parameters is ensured to be less than or equal to 5ms, the acquisition frequency can be dynamically reduced to 50Hz aiming at slowly-variable parameters such as the roughness of a riveting surface, the acquisition result is temporarily stored through the data caching mechanism, and the data is synchronously written into a database after the acquisition of all the parameters is completed, so that the data mismatch caused by single parameter acquisition delay is avoided. Calibrating a reference spin riveting gap of a rotary fastener, establishing a multi-dimensional parameter, reference gap and load level association database of the fastener, embedding an automatic recognition function of the type of the fastener, realizing the integrated adaptation of pre-detection, automatic calibration of the reference gap and load level matching of the fastener to be spin riveted, collecting initial wear data of a tool and an environmental temperature reference value, and recording the initial wear data and the environmental temperature reference value into the database. Preferably, the model construction stage combines