Rapid locking method and locking structure for automobile part mold

Abstract

The invention relates to a quick locking method and structure for an automobile part mold. The device is characterized by comprising a fixed base, a locking cylinder and a locking unit, in the motion control of the locking mechanism, a motion smoothing algorithm and a trajectory optimization algorithm are introduced, the motion trajectory of the locking mechanism is smoothed, the speed change and the acceleration change in the motion process are reduced, the vibration and impact of the locking mechanism are reduced, and the stability of the locking process is improved, so that the consistency of the locking position of the automobile part mold is ensured; and after the motion control is optimized, transmitting the optimized motion track data to a machine vision system for monitoring the subsequent locking process.

Description

Technical Field

[0001] The invention relates to the technical field of automobile production, and in particular to a method for quickly locking an automobile component mold and a locking structure thereof. Background Art

[0002] In the rapid locking method of automobile parts molds, the movement and positioning of automobile parts molds during the locking process are crucial to the locking consistency. However, due to the differences in the shapes and sizes of automobile parts molds, they are prone to offset or shaking during the locking process, resulting in deviations in the locking position and force, which in turn affects product quality. This problem is particularly prominent in mass production, because even small deviations may accumulate and amplify as the production quantity increases, ultimately leading to a serious decline in product quality.

[0003] In addition, during the locking process, the automotive component mold may be disturbed by external vibration and impact, further exacerbating the instability of its movement and positioning. These interference factors may come from mechanical vibration, air flow disturbance, etc. in the production environment, or from the movement and impact of the locking mechanism itself. How to achieve accurate positioning and stable locking of automotive component molds in a complex production environment is a technical problem that needs to be solved urgently.

[0004] At the same time, the rapid locking method of automobile parts molds also faces the contradiction between efficiency and precision. On the one hand, in order to improve production efficiency, the locking time needs to be shortened as much as possible; on the other hand, in order to ensure the locking quality, the automobile parts mold needs to be accurately positioned and stably clamped during the locking process. How to maximize the locking efficiency while ensuring the locking consistency is another technical issue worthy of in-depth study. Summary of the invention

[0005] The technical problem to be solved by the present invention is to provide a method for quickly locking a mold of an automobile part and a locking structure thereof, which can solve the problem of the contradiction between efficiency and precision faced by the traditional method for quickly locking a mold of an automobile part.

[0006] In order to solve the above technical problems, the technical solution of the present invention is: a method for quickly locking a mold of an automobile part, and its innovation lies in: the specific locking method is as follows: S1: Acquire the 3D model data of the automobile component mold to be locked, determine the shape characteristics and size parameters of the automobile component mold, establish a corresponding data model of the locking position and locking force, collect the position and posture information of the automobile component mold, and make corresponding compensation adjustments by controlling the locking mechanism; S2: Use sensors to obtain vibration and airflow parameters in the production environment, establish a correlation model between interference factors in the production environment and locking accuracy, set up vibration reduction devices and buffer devices on the locking mechanism, and dynamically adjust the motion control strategy of the locking mechanism based on the vibration and airflow parameters collected in real time; S3: In the motion control of the locking mechanism, the motion smoothing algorithm and trajectory optimization algorithm are introduced to smooth the motion trajectory of the locking mechanism, reduce the speed change and acceleration change during the motion process, reduce its own vibration and impact, improve stability, and transmit the optimized motion trajectory data to the machine vision system for subsequent locking process monitoring; S4: Use high-speed industrial cameras and deep learning image processing algorithms for real-time monitoring. By processing and analyzing the collected image data, the locking position and locking status of the automotive component mold can be determined in real time. The locking position and status information obtained by the machine vision system can be compared with the optimized motion trajectory data. S5: Using the results obtained in the previous steps, a data association model between locking parameters and product quality is established. Through support vector machine and random forest machine learning algorithms, the association rules between locking parameters and product quality are explored, and the setting of locking parameters is optimized for different automotive component molds and production conditions.

[0007] A locking structure for a mold of an automobile part, the innovation of which is that it comprises a fixed base, a locking cylinder and a locking unit; The fixed base comprises a bottom plate, a cylinder fixing plate and a lock hinge seat; the bottom plate is a rectangular plate structure, a cylinder fixing plate is arranged on the upper surface of the bottom plate near one end, and a lock hinge seat is arranged on the upper surface of the bottom plate near the other end; a locking cylinder fixing hole and a through hole for accommodating the output end of the locking cylinder are opened on the cylinder fixing plate; a lock hinge hole is arranged on the lock hinge seat; a slide rail is arranged on the upper surface of the bottom plate between the cylinder fixing plate and the lock hinge seat, and a slider is arranged on the slide rail; The locking cylinder is mounted on the cylinder fixing plate in parallel with the base plate; The locking unit is mounted on the output end of the locking cylinder; the locking unit comprises a lock head connecting seat and a lock head; the lock head connecting seat is mounted on the output end of the locking cylinder, the lock head connecting seat is in a U-shaped structure, the opening direction of the lock head connecting seat is parallel to the bottom plate, and the upper and lower side surfaces of the lock head connecting seat are connected; a guide pin is horizontally arranged at a position close to the opening end on the inner side wall of the lock head connecting seat; the lower side surface of the lock head connecting seat is mounted on the slider; The lock is in the form of a rectangular plate, which is embedded in a lock connection seat and has one end extending out of the lock connection seat; a hinged plate is provided on the bottom side of the lock, and the hinged plate is mounted on the lock hinge seat through a hinge shaft; an arc-shaped guide groove is provided on the lock, and a guide pin on the lock connection seat is embedded in the arc-shaped guide groove on the lock; a mold locking bayonet is provided on one end of the lock extending out of the lock connection seat; the lock connection seat is driven to reciprocate along the extension direction of the slide rail by a locking cylinder, and the guide pin moves along the arc-shaped guide groove of the lock to drive the lock to rotate around the hinge, so that the mold locking bayonet at the end of the lock can lock or release the automobile part mold.

[0008] The advantages of the present invention are: 1) In the present invention, in order to solve the accuracy and consistency problems in the locking process of automobile parts molds, the present invention first obtains the three-dimensional model data of the parts and establishes a data model of the locking position and force. During the locking process, the position and posture information of the parts is collected in real time and compared with the preset position, and compensation adjustment is performed when an offset is found. At the same time, the environmental vibration and airflow parameters are obtained through sensors, and a correlation model between interference factors and locking accuracy is established, and a vibration reduction buffer device and an adaptive compensation strategy are used to deal with external interference. In addition, the present invention introduces a motion smoothing algorithm to optimize the trajectory of the locking mechanism, and uses a machine vision system to monitor the locking state in real time. Finally, based on the various types of data collected, an association model between locking parameters and product quality is established, and the parameter settings are optimized through a machine learning algorithm. The present invention realizes the intelligent and refined control of the locking process of automobile parts molds, effectively improves the locking accuracy and consistency, and ensures product quality. BRIEF DESCRIPTION OF THE DRAWINGS

[0009] The present invention will be further described in detail below in conjunction with the accompanying drawings and specific embodiments.

[0010] Figure 1 The present invention is a flow chart of a method for quickly locking a mold for an automobile part.

[0011] Figure 2 The figure is a schematic diagram of a method for quickly locking a mold for an automobile part according to the present invention.

[0012] Figure 3 The present invention is another schematic diagram of a method for quickly locking a mold for an automobile part.

[0013] Figure 4 The present invention is a locking structure diagram of a method for quickly locking a mold for an automobile part.

[0014] Figure 5 The present invention is a partial structural diagram of the locking structure of a method for quickly locking a mold for an automobile part. DETAILED DESCRIPTION

[0015] In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Generally, the components of the embodiments of the present invention described and shown in the drawings here can be arranged and designed in various different configurations.

[0016] Therefore, the following detailed description of the embodiments of the present invention provided in the accompanying drawings is not intended to limit the scope of the invention claimed for protection, but merely represents selected embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0017] like Figures 1 to 3 A method for quickly locking a mold for an automobile part is shown; The three-dimensional model data of the automobile part mold to be locked is obtained, and the shape characteristics and size parameters of the automobile part mold are determined by analyzing the three-dimensional model data. According to the shape characteristics and size parameters of the automobile part mold, a data model of the corresponding locking position and locking force is established. During the locking process, the posture information of the automobile part mold is collected in real time, and the collected posture information is compared with the preset locking position. By comparing the posture information with the preset locking position, it is determined whether the automobile part mold is offset or shaken. If the judgment result is that the automobile part mold is offset or shaken, the direction and amplitude information of the offset or shake is obtained. According to the acquired offset or shake direction and amplitude information, the locking mechanism is controlled to make corresponding compensation adjustments. Through the compensation adjustment of the locking mechanism, the final locking position of the automobile part mold is ensured to be accurate and correct, and the high-precision automobile part mold locking operation is completed.

[0018] Exemplarily, first, a three-dimensional scanner is used to scan the automobile part mold to be locked to obtain its three-dimensional model data. By analyzing the three-dimensional model data, the key shape features of the automobile part mold, such as plane, cylindrical surface, conical surface, etc., are extracted, and the size parameters of each feature, such as length, width, height, diameter, etc., are calculated. According to the extracted shape features and size parameters, a mathematical model is established to describe the locking position and required locking force of the automobile part mold. For example, for a rectangular automobile part mold, its mathematical model can be represented by a point and three vectors in a three-dimensional coordinate system, the point represents the locking position, and the vector represents the locking force in the length, width, and height directions. During the locking process, the posture information of the automobile part mold is collected in real time by a visual sensor, and compared with the preset locking position in the mathematical model. During the comparison, a posture estimation algorithm, such as an ICP algorithm, can be used to calculate the deviation between the current posture of the automobile part mold and the preset posture. If the deviation exceeds a preset threshold (such as 2mm), it is determined that an offset or shake has occurred. At the same time, by analyzing the direction and size of the posture deviation, the specific direction and amplitude of the offset or shake can be calculated. For example, if the deviation in the X direction is 3mm and the deviation in the Y direction is -1mm, the offset direction is the positive X direction and the negative Y direction, and the amplitude is sqrt(3^2+1^2)=16mm. According to the calculated direction and amplitude of the offset or shake, the locking mechanism is controlled to make compensation adjustments, and the adjustment amount is equal to the offset amount but in the opposite direction. For example, for the above-mentioned offset situation, the locking mechanism needs to move 3mm in the negative X direction and 1mm in the positive Y direction, so as to adjust the automotive component mold back to the preset locking position. Through this real-time compensation adjustment, it can be ensured that the position accuracy of the automotive component mold after locking is completed is controlled within 1mm, and a high-precision locking operation can be achieved.

[0019] The vibration and airflow parameter data collected by the acceleration sensor and the airflow sensor are obtained, and the vibration and airflow parameter data are input into the pre-established production environment interference factor and locking accuracy association model to obtain the estimated locking accuracy under the current production environment; it is determined whether the estimated locking accuracy meets the preset accuracy threshold. If the threshold is met, the locking mechanism is controlled to perform the locking operation according to the standard motion control strategy; if the estimated accuracy does not meet the threshold, the vibration reduction device and the buffer device are triggered to work; the vibration reduction device adjusts the damping coefficient in real time according to the vibration parameter size, actively reduces the vibration of the locking mechanism, and attenuates the influence of external vibration on the locking accuracy; the buffer device uses elastic elements to absorb Instantaneous impact force is prevented from directly acting on the locking mechanism, causing the position displacement of the automobile component mold; after vibration reduction and buffering treatment, the acceleration sensor and airflow sensor data are obtained again, and input into the associated model to re-estimate the locking accuracy; if the estimated accuracy still does not meet the standard, the movement speed, acceleration, torque and other parameters of the locking mechanism are dynamically adjusted according to the current vibration and airflow parameters, and the locking accuracy is further improved through the adaptive compensation locking control strategy; after the locking is completed, the actual locking accuracy data is obtained and input into the machine learning algorithm to conduct online training and optimization of the associated model, so as to improve the estimation accuracy of the associated model and provide an optimization basis for subsequent locking operations.

[0020] Exemplarily, during the locking operation, the acceleration sensor and the airflow sensor collect vibration and airflow parameter data in real time, with a sampling frequency of 1000Hz. The collected data is input into the production environment interference factor and locking accuracy association model established based on the support vector machine algorithm. Through Gaussian kernel function mapping and sequence minimum optimization algorithm, the estimated locking accuracy under the current production environment is obtained within 10ms. The system determines whether the estimated locking accuracy meets the preset accuracy threshold of ±1mm. If the threshold is met, the locking mechanism is controlled to perform standard locking operations according to the preset parameters such as movement speed 5m / s, acceleration 2m / s², and torque 10N·m. If the estimated accuracy does not meet the threshold, the vibration reduction device and the buffer device are triggered to work. The vibration reduction device uses a magnetorheological damper. According to the vibration frequency and amplitude, the damping coefficient of 0~50N·s / mm is adjusted in real time through the fuzzy PID control algorithm to actively reduce the vibration of the locking mechanism and attenuate more than 50% of the external vibration effect. The buffer device uses a memory alloy spring. When an instantaneous impact force greater than 10N is detected, the spring automatically compresses and deforms to absorb 80% of the impact energy, preventing the impact force from directly acting on the locking mechanism and causing the position and posture of the automotive component mold to shift. After vibration reduction and buffering, the system obtains sensor data again and estimates the locking accuracy. If the accuracy still does not meet the requirement of ±1mm, the movement speed, acceleration, torque and other parameters of the locking mechanism are dynamically adjusted through the adaptive neural network algorithm according to the deviation between the current vibration and airflow parameters and the preset template to achieve adaptive compensation control, further improving the locking accuracy to within ±05mm. After the locking is completed, the system obtains the actual locking accuracy data measured by the laser displacement sensor and inputs it into the machine learning algorithm based on the long short-term memory network. The association model is trained and optimized online through the back propagation algorithm to continuously improve the estimated accuracy of the association model, so that the error between the estimated value and the measured value is reduced to within 5%, providing a precise optimization basis for subsequent locking operations.

[0021] According to the motion control requirements of the locking mechanism, the motion trajectory of the locking mechanism is smoothed by a motion smoothing algorithm to obtain smoothed motion trajectory data. According to the smoothed motion trajectory data, the motion trajectory of the locking mechanism is optimized by a trajectory optimization algorithm to obtain optimized motion trajectory data, and the optimized motion trajectory data contains speed change and acceleration change information. According to the optimized motion trajectory data, the locking mechanism is controlled to move according to the optimized motion trajectory, and the vibration impact during the movement of the locking mechanism is reduced by reducing the speed change and acceleration change. During the locking process, the motion parameters of the locking mechanism, including speed, acceleration, etc., are obtained in real time to determine whether the motion parameters meet the preset stability requirements. If not, the motion control parameters of the locking mechanism are adjusted. The locking position information of the automotive component mold is obtained, and the locking position information is compared with the preset standard locking position to determine whether the locking position meets the consistency requirements. If not, the motion control parameters of the locking mechanism are adjusted. The optimized motion trajectory data is transmitted to the machine vision system. Based on the motion trajectory data, the machine vision system monitors the locking process in real time, obtains the actual motion trajectory of the locking mechanism, and determines whether the actual motion trajectory is consistent with the optimized motion trajectory. If not, an alarm signal is generated. Based on the monitoring results of the machine vision system, a machine learning algorithm is used to adaptively optimize the motion control parameters of the locking mechanism, including motion smoothing algorithm parameters and trajectory optimization algorithm parameters, to continuously improve the stability of the locking process and the consistency of the locking position.

[0022] Exemplarily, according to the motion control requirements of the locking mechanism, the motion trajectory of the locking mechanism is smoothed by using a quintic polynomial interpolation algorithm. By selecting appropriate trajectory sampling points, such as 1s intervals, the least squares method is used to fit the smoothed motion trajectory curve equation so that the speed and acceleration change continuously. Then the gradient descent algorithm is used to optimize the smoothed motion trajectory. By setting constraints such as the speed change threshold ±05m / s and the acceleration change threshold ±5m / s², the trajectory parameters are iteratively optimized to reduce the speed standard deviation by 20% and the acceleration standard deviation by 30%, and the optimal motion trajectory is obtained. The locking mechanism is controlled to move according to the optimized trajectory, and the motion parameters are collected in real time. If the speed exceeds ±03m / s and the acceleration exceeds ±3m / s², the PID algorithm adjusts the servo motor speed to make the movement smoother. The locking position image is collected by machine vision, the edge detection algorithm is used to extract the contour, and the pixel coordinate deviation is calculated. If the deviation exceeds ±2mm, the feedback is adjusted to adjust the end point position of the movement. The optimized trajectory is transmitted to the visual system for real-time comparison and monitoring. If the actual trajectory deviates from the optimized trajectory by more than ±1mm, an alarm is issued. Based on monitoring data, a neural network algorithm is used for self-learning optimization. By setting 1,000 sets of training samples and adjusting the number of hidden layer nodes in the network to 50, the training error is reduced to below 5%, making the locking trajectory more accurate and controllable, and realizing intelligent locking.

[0023] High-speed industrial cameras and deep learning image processing algorithms are used to monitor the locking process in real time. The locking position and locking status of automotive component molds are determined in real time by processing and analyzing the collected image data. The locking position and status information obtained by the machine vision system are compared with the optimized motion trajectory data. Once an abnormal deviation is found, the alarm mechanism is immediately triggered, and the locking parameters are automatically adjusted according to the deviation to ensure that the locking quality meets the requirements.

[0024] Step 1: A series of image data in the locking process of the automotive component mold is collected in real time by a high-speed industrial camera to obtain a continuous image sequence of the locking process. Step 2: The collected image data is input into a pre-trained deep learning model, which can be a convolutional neural network or a recursive neural network. Through feature extraction and classification recognition, the locking position coordinates and locking state category of the automotive component mold in each frame of the image are determined. Step 3: According to the recognition results of the deep learning model, the real-time position change and state change of the automotive component mold during the locking process are obtained to form time series data reflecting the locking process. Step 4: The locking position and state time series data obtained by machine vision are compared with the standard locking motion trajectory optimized according to the material characteristics, external dimensions and other parameters of the automotive component mold, and the similarity of the two time series is calculated by the dynamic time warping algorithm. Step 5: Set the locking position deviation threshold and the locking state abnormality threshold. When the similarity is lower than the threshold, it is judged as a locking deviation or locking abnormality, triggering an audible and visual alarm prompt, and the deviation data is fed back to the control system. Step 6: The control system uses a parameter adaptive optimization algorithm, such as a particle swarm optimization algorithm, based on the locking deviation data to adjust the operating parameters of the locking device, such as the locking force and locking time, so that the locking process returns to the optimal state. Step 7: Continue to perform machine vision monitoring on the adjusted locking process, and enter the locking data into the sample set for continuous training and improvement of the deep learning model and motion trajectory optimization algorithm, and continuously improve the accuracy and stability of locking quality detection and control.

[0025] Exemplarily, a series of image data of the locking process of the automobile parts mold is collected in real time by a high-speed industrial camera. The camera sampling frequency is set to 200 frames / second, the resolution is 1920×1080 pixels, and the locking process is continuously collected for 10 seconds to obtain a 2000-frame image sequence. The collected image data is input into a pre-trained convolutional neural network model, which contains 5 convolutional layers and 3 fully connected layers. Through feature extraction and classification recognition, the locking position coordinates (x, y) and locking status category (0 for unlocked, 1 for locked) of the automobile parts mold in each frame of the image are judged with an accuracy of 98%. According to the recognition results of the convolutional neural network model, the real-time position change and state change of the automobile parts mold during the locking process are obtained to form a time series containing 2000 (x, y, status) data points. The time series is compared with the standard locking motion trajectory (including 2000 standard position coordinates and state values) optimized according to the material characteristics, external dimensions and other parameters of the automotive parts mold. The similarity of the two time series is calculated by the dynamic time warping algorithm. When the similarity is lower than 95, it is judged as locking deviation or locking abnormality, triggering an audible and visual alarm prompt, and the deviation data (the difference between the actual value and the standard value) is fed back to the control system. The control system uses the particle swarm optimization algorithm to adaptively adjust the operating parameters of the locking device according to the deviation data, such as increasing the locking force from 1000N to 1050N and extending the locking time from 5s to 2s, so that the locking process returns to the optimal state after 3 iterations (similarity greater than or equal to 95). The adjusted locking process continues to be monitored in real time by machine vision, and the 2000 locking data points are entered into the sample set for incremental training of the convolutional neural network model and dynamic correction of the fitness function of the particle swarm optimization algorithm, so as to continuously improve the accuracy and stability of locking quality detection and control.

[0026] The three-dimensional model data of the automotive parts mold, including its geometric dimensions, material properties and other information, is obtained and converted into a unified digital representation to lay the foundation for subsequent analysis. The position and posture information of the automotive parts mold on the production line is collected through sensors to determine whether it meets the locking process requirements. If not, it is adjusted in time to ensure the accuracy of the locking position. The temperature, humidity, vibration and other parameters of the production environment are obtained to analyze their impact on the locking quality. When the environmental parameters exceed the preset threshold, the locking parameters are adjusted in time or corresponding measures are taken to reduce the interference of environmental factors. The locking process is monitored in real time by machine vision technology to obtain key parameters such as locking torque and locking angle to determine whether the locking quality meets the requirements. If an abnormality occurs, an alarm is issued in time and the cause is analyzed. According to the locking parameters and product quality test results, machine learning algorithms such as support vector machines and random forests are used to establish a nonlinear correlation model between locking parameters and product quality to explore the intrinsic relationship between the two. For different models of automotive parts molds and production conditions, the optimal locking parameter combination is analyzed through the correlation model to improve the locking efficiency while ensuring product quality and realize the dynamic optimization of locking process parameters. The optimized locking parameter settings are applied to the production line. The locking process and product quality are continuously monitored through technologies such as machine vision. If deviations occur, timely feedback and parameter adjustments are made to form a closed-loop control, thus achieving intelligent and refined management and control of the locking process of automotive component molds.

[0027] For example, first use a 3D scanner to scan the automotive component mold to obtain its 3D point cloud data. The point cloud density is set to 1mm and the scanning accuracy reaches 0.5mm. Then use Geomagic Design X software to post-process the point cloud data, convert the point cloud into a NURBS surface model through the surface reconstruction algorithm, and extract its geometric size information. At the same time, use a material composition analyzer to determine the material composition of the automotive component mold, and add material property information to the 3D model in combination with the material mechanical properties test data. Next, install a visual sensor on the production line to collect the position and posture information of the automotive component mold, and use the posture estimation algorithm to determine whether its deviation is within 1°. If it exceeds, it will be automatically adjusted by the robotic arm. For environmental parameters such as temperature and humidity, high-precision sensors are used to collect them in real time. When the temperature exceeds 30°C or the humidity exceeds 70%, an alarm is automatically triggered to adjust the workshop air conditioning and dehumidification equipment. During the locking process, the machine vision system monitors the locking torque and angle in real time at a speed of 50 frames per second, and determines whether the torque is within the standard range through the threshold segmentation algorithm. If the deviation exceeds 5%, an alarm is triggered. The locking parameters and quality inspection results are input into the machine learning model, and the nonlinear association model of locking parameters and product quality is established by the support vector machine algorithm. The model hyperparameters are optimized through grid search and cross-validation, and the model accuracy reaches more than 95%. Finally, the genetic algorithm is used to search for the optimal locking parameter combination, with product quality and locking efficiency as the optimization goals. After 100 iterations, the optimal parameter combination is obtained and applied to the production line. The locking process is continuously monitored through a closed-loop control system. If the quality inspection results are out of tolerance for three consecutive times, the locking parameters are automatically adjusted to ensure stable product quality.

[0028] like Figures 4 to 5 A locking structure for an automobile component mold comprises a fixed base 1, a locking cylinder 2 and a locking unit 3; The fixed base 1 includes a bottom plate 11, a cylinder fixing plate 12 and a lock hinge seat 13; the bottom plate 11 is a rectangular plate structure, the cylinder fixing plate 12 is arranged near one end of the upper surface of the bottom plate 11, and the lock hinge seat 13 is arranged near the other end of the upper surface of the bottom plate 11; the cylinder fixing plate 12 is provided with a locking cylinder fixing hole and a through hole for accommodating the output end of the locking cylinder 2 to pass through; the lock hinge seat 13 is provided with a lock hinge hole; a slide rail 14 is arranged on the upper surface of the bottom plate 11 between the cylinder fixing plate 12 and the lock hinge seat 13, and a slider is arranged on the slide rail 14; The locking cylinder 2 is mounted on the cylinder fixing plate in parallel with the base plate 11; The locking unit 3 is mounted on the output end of the locking cylinder 2; the locking unit 3 includes a lock head connection seat 31 and a lock head 32; the lock head connection seat 31 is mounted on the output end of the locking cylinder 2, the lock head connection seat 31 is a U-shaped structure, the opening direction of the lock head connection seat 31 is parallel to the bottom plate 11, and the upper and lower side surfaces of the lock head connection seat 31 are connected; a guide pin 33 is horizontally arranged on the inner side wall of the lock head connection seat 31 near the opening end; the lower side surface of the lock head connection seat 31 is mounted on the slider; The lock 32 is a rectangular plate-shaped structure, which is embedded in the lock connecting seat 31 and one end of which extends out of the lock connecting seat 31; a hinged plate is provided on the bottom side of the lock, and the hinged plate is installed on the lock hinge seat 13 through a hinge shaft; an arc-shaped guide groove 34 is opened on the lock 32, and the guide pin on the lock connecting seat is embedded in the arc-shaped guide groove 34 on the lock; a mold locking bayonet 35 is provided on the end of the lock extending out of the lock connecting seat 31; the lock connecting seat 31 is driven to reciprocate along the extension direction of the slide rail 14 by the locking cylinder 2, and the guide pin 33 moves along the arc-shaped guide groove 34 of the lock to drive the lock 32 to rotate around the hinge, so that the mold locking bayonet 35 at the end of the lock 32 locks or releases the automobile part mold.

[0029] Those skilled in the art should understand that the present invention is not limited to the above embodiments, and the above embodiments and descriptions are only for explaining the principles of the present invention. Without departing from the spirit and scope of the present invention, the present invention may have various changes and improvements, and these changes and improvements fall within the scope of the present invention to be protected. The scope of protection of the present invention is defined by the attached claims and their equivalents.

Claims

1. A method for quickly locking a mold for an automobile part, characterized in that: The specific locking method is as follows: S1: Acquire the 3D model data of the automobile component mold to be locked, determine the shape characteristics and size parameters of the automobile component mold, establish a corresponding data model of the locking position and locking force, collect the position and posture information of the automobile component mold, and make corresponding compensation adjustments by controlling the locking mechanism; S2: Use sensors to obtain vibration and airflow parameters in the production environment, establish a correlation model between interference factors in the production environment and locking accuracy, set up vibration reduction devices and buffer devices on the locking mechanism, and dynamically adjust the motion control strategy of the locking mechanism based on the vibration and airflow parameters collected in real time; S3: In the motion control of the locking mechanism, the motion smoothing algorithm and trajectory optimization algorithm are introduced to smooth the motion trajectory of the locking mechanism, reduce the speed change and acceleration change during the motion process, reduce its own vibration and impact, improve stability, and transmit the optimized motion trajectory data to the machine vision system for subsequent locking process monitoring; S4: Use high-speed industrial cameras and deep learning image processing algorithms for real-time monitoring. By processing and analyzing the collected image data, the locking position and locking status of the automotive component mold can be determined in real time. The locking position and status information obtained by the machine vision system can be compared with the optimized motion trajectory data. S5: Using the results obtained in the previous steps, a data association model between locking parameters and product quality is established. Through support vector machine and random forest machine learning algorithms, the association rules between locking parameters and product quality are explored, and the setting of locking parameters is optimized for different automotive component molds and production conditions.

2. The method for quickly locking a mold for an automobile part according to claim 1, characterized in that: The specific method in S1 includes: Acquire the three-dimensional model data of the automobile component mold to be locked, and determine the shape characteristics and size parameters of the automobile component mold by analyzing the three-dimensional model data; According to the shape characteristics and size parameters of the automobile parts mold, a data model of the corresponding locking position and locking force is established; During the locking process, the position information of the automobile component mold is collected in real time, and the collected position information is compared with the preset locking position; By comparing the posture information with the preset locking position, it is determined whether the automotive component mold is offset or shaking; If the judgment result is that the automobile component mold is offset or shaken, the direction and amplitude information of the offset or shake is obtained; According to the acquired information on the direction and amplitude of the offset or shaking, the locking mechanism is controlled to make corresponding compensation adjustments; Through the compensation adjustment of the locking mechanism, the final locking position of the automobile component mold is ensured to be accurate and correct, completing the high-precision locking operation of the automobile component mold.

3. The method for quickly locking a mold for an automobile part according to claim 1, characterized in that: The specific method in S2 includes: Obtain the vibration and airflow parameter data collected by the acceleration sensor and the airflow sensor, input the vibration and airflow parameter data into the pre-established production environment interference factor and locking accuracy association model, and obtain the estimated locking accuracy under the current production environment; Determine whether the estimated locking accuracy meets a preset accuracy threshold, and if so, control the locking mechanism to perform a locking operation according to a standard motion control strategy; If the estimated accuracy does not meet the threshold, the vibration reduction device and the buffer device are triggered to work; The vibration reduction device adjusts the damping coefficient in real time according to the vibration parameters, actively reduces the vibration of the locking mechanism, and attenuates the influence of external vibration on the locking accuracy; The buffer device uses elastic elements to absorb instantaneous impact force, preventing the impact force from directly acting on the locking mechanism and causing the position deviation of the automotive component mold; After the vibration reduction and buffering process, the acceleration sensor and airflow sensor data are obtained again and input into the correlation model to re-estimate the locking accuracy; If the estimated accuracy still does not meet the standard, the movement speed, acceleration, torque and other parameters of the locking mechanism are dynamically adjusted according to the current vibration and airflow parameters, and the locking accuracy is further improved through the adaptive compensation locking control strategy; After locking is completed, the actual locking accuracy data is obtained and input into the machine learning algorithm to perform online training and optimization on the associated model, thereby improving the estimated accuracy of the associated model and providing an optimization basis for subsequent locking operations.

4. The method for quickly locking a mold for an automobile part according to claim 1, characterized in that: The specific method in S3 includes: According to the motion control requirements of the locking mechanism, a motion smoothing algorithm is used to smooth the motion trajectory of the locking mechanism to obtain smoothed motion trajectory data; According to the smoothed motion trajectory data, the motion trajectory of the locking mechanism is optimized by using a trajectory optimization algorithm to obtain optimized motion trajectory data, wherein the optimized motion trajectory data includes speed change and acceleration change information; According to the optimized motion trajectory data, the locking mechanism is controlled to move according to the optimized motion trajectory, and the vibration impact during the movement of the locking mechanism is reduced by reducing the speed change and acceleration change; During the locking process, the motion parameters of the locking mechanism, including speed and acceleration, are obtained in real time to determine whether the motion parameters meet the preset stability requirements. If not, the motion control parameters of the locking mechanism are adjusted; Obtain the locking position information of the automobile component mold, compare the locking position information with the preset standard locking position, and determine whether the locking position meets the consistency requirements. If not, adjust the motion control parameters of the locking mechanism; The optimized motion trajectory data is transmitted to the machine vision system. The machine vision system monitors the locking process in real time based on the motion trajectory data, obtains the actual motion trajectory of the locking mechanism, and determines whether the actual motion trajectory is consistent with the optimized motion trajectory. If not, an alarm signal is generated; According to the monitoring results of the machine vision system, a machine learning algorithm is used to adaptively optimize the motion control parameters of the locking mechanism, including motion smoothing algorithm parameters and trajectory optimization algorithm parameters, to continuously improve the smoothness of the locking process and the consistency of the locking position.

5. The method for quickly locking a mold for an automobile part according to claim 1, characterized in that: The specific method in S4 includes: Step 1: A series of image data of the locking process of the automobile component mold is collected in real time by a high-speed industrial camera to obtain a continuous image sequence of the locking process; Step 2: Input the collected image data into a pre-trained deep learning model, which can be a convolutional neural network or a recursive neural network, and determine the locking position coordinates and locking state category of the automotive component mold in each frame of the image through feature extraction and classification recognition; Step 3: According to the recognition results of the deep learning model, the real-time position change and state change of the automobile component mold during the locking process are obtained to form time series data reflecting the locking process; Step 4: Compare the locking position and state time series data obtained by machine vision with the standard locking motion trajectory optimized according to parameters such as the material characteristics and external dimensions of the automotive component mold, and calculate the similarity of the two time series through the dynamic time warping algorithm; Step 5: Set the locking position deviation threshold and the locking state abnormality threshold. When the similarity is lower than the threshold, it is judged as locking deviation or locking abnormality, triggering an audible and visual alarm prompt, and feeding back the deviation data to the control system; Step 6: The control system uses a parameter adaptive optimization algorithm, such as a particle swarm optimization algorithm, according to the locking deviation data to adjust the operating parameters of the locking device, such as the locking force and the locking time, so that the locking process is restored to the optimal state; Step 7: Continue to perform machine vision monitoring on the adjusted locking process, and enter the locking data into the sample set for continuous training and improvement of the deep learning model and motion trajectory optimization algorithm, so as to continuously improve the accuracy and stability of locking quality detection and control.

6. The method for quickly locking a mold for an automobile part according to claim 1, characterized in that: The specific method in S5 includes: Obtain 3D model data of automotive component molds, including information such as their geometric dimensions and material properties, and convert them into a unified digital representation to lay the foundation for subsequent analysis; The position and posture information of the automobile parts mold on the production line is collected through sensors to determine whether it meets the locking process requirements. If not, it is adjusted in time to ensure the accuracy of the locking position; Obtain parameters such as temperature, humidity and vibration of the production environment, analyze their impact on the locking quality, and when environmental parameters exceed the preset thresholds, promptly adjust the locking parameters or take corresponding measures to reduce the interference of environmental factors; Use machine vision technology to monitor the locking process in real time, obtain key parameters such as locking torque and locking angle, and determine whether the locking quality meets the requirements. If any abnormality occurs, promptly alarm and analyze the cause; According to the locking parameters and product quality test results, machine learning algorithms such as support vector machine and random forest are used to establish a nonlinear correlation model between locking parameters and product quality, and explore the intrinsic relationship between the two; According to different models of automobile parts molds and production conditions, the optimal locking parameter combination is analyzed through the correlation model, which improves the locking efficiency while ensuring product quality and realizes the dynamic optimization of locking process parameters; The optimized locking parameter settings are applied to the production line. The locking process and product quality are continuously monitored through technologies such as machine vision. If deviations occur, timely feedback and parameter adjustments are made to form a closed-loop control, thus achieving intelligent and refined management and control of the locking process of automotive component molds.

7. An automobile component mold locking structure, characterized in that: It includes a fixed base, a locking cylinder and a locking unit; The fixed base comprises a bottom plate, a cylinder fixing plate and a lock hinge seat; the bottom plate is a rectangular plate structure, a cylinder fixing plate is arranged on the upper surface of the bottom plate near one end, and a lock hinge seat is arranged on the upper surface of the bottom plate near the other end; a locking cylinder fixing hole and a through hole for accommodating the output end of the locking cylinder are opened on the cylinder fixing plate; a lock hinge hole is arranged on the lock hinge seat; a slide rail is arranged on the upper surface of the bottom plate between the cylinder fixing plate and the lock hinge seat, and a slider is arranged on the slide rail; The locking cylinder is mounted on the cylinder fixing plate in parallel with the base plate; The locking unit is mounted on the output end of the locking cylinder; the locking unit comprises a lock head connecting seat and a lock head; the lock head connecting seat is mounted on the output end of the locking cylinder, the lock head connecting seat is in a U-shaped structure, the opening direction of the lock head connecting seat is parallel to the bottom plate, and the upper and lower side surfaces of the lock head connecting seat are connected; a guide pin is horizontally arranged at a position close to the opening end on the inner side wall of the lock head connecting seat; the lower side surface of the lock head connecting seat is mounted on the slider; The lock is in the form of a rectangular plate, which is embedded in a lock connection seat and has one end extending out of the lock connection seat; a hinged plate is provided on the bottom side of the lock, and the hinged plate is mounted on the lock hinge seat through a hinge shaft; an arc-shaped guide groove is provided on the lock, and a guide pin on the lock connection seat is embedded in the arc-shaped guide groove on the lock; a mold locking bayonet is provided on one end of the lock extending out of the lock connection seat; the lock connection seat is driven to reciprocate along the extension direction of the slide rail by a locking cylinder, and the guide pin moves along the arc-shaped guide groove of the lock to drive the lock to rotate around the hinge, so that the mold locking bayonet at the end of the lock can lock or release the automobile part mold.

Images