A die-casting mold stripping and part-taking adaptive control system and method

Abstract

This invention belongs to the field of die casting molding technology, specifically an adaptive control system and method for demolding and part removal in die casting molds. The system includes a die casting machine, a sensing module, a control center, an execution module, and an emergency stop button mounted on the surface of the die casting machine. The sensing module includes vision sensors, pressure sensors, temperature sensors, laser sensors, vibration sensors, and ultrasonic thickness sensors. The control center uses an industrial control computer running a real-time operating system, including the following functions: A3. Adaptive Algorithm: including PID control and a machine learning model. This adaptive control system and method for demolding and part removal in die casting molds, through an adaptive PID algorithm and a machine learning model, can dynamically adjust the spraying pressure, time, and amount of release agent in the demolding device based on the mold temperature, residue on the mold, and mold wear monitored in real time by the sensing module. This ensures optimal demolding results while saving resources and avoiding mold damage.

Description

Technical Field

[0001] This invention relates to the field of die casting molding technology, and in particular to an adaptive control system and method for demolding and removing parts from die casting molds. Background Technology

[0002] In the current die-casting production process, mold demolding and part removal largely rely on manual operation or semi-automatic equipment, which presents the following prominent problems:

[0003] 1. Low efficiency: The slow pace of manual demolding and part removal makes it difficult to match the high-speed die-casting cycle, becoming a production bottleneck;

[0004] 2. Unstable quality: Manual judgment of demolding effect and part removal force can easily lead to incomplete demolding, damage to parts, or scratches on the mold;

[0005] 3. Poor adaptability: Automated equipment with fixed parameters cannot dynamically adjust according to changes in mold temperature, residue condition, mold wear, etc., resulting in poor performance when working conditions change;

[0006] 4. Waste of resources: Lack of precise control over the amount of release agent and demolding time can easily lead to waste;

[0007] 5. High maintenance costs: The condition of the mold and the clogging of the filter screen rely on manual inspection. Failure to detect faults in time may exacerbate equipment wear and tear.

[0008] Therefore, this application proposes an adaptive control system and method for demolding and part removal of die-casting molds to achieve efficient, accurate and reliable mold demolding and part removal operations. Summary of the Invention

[0009] Based on the aforementioned technical problems, this invention proposes an adaptive control system and method for demolding and part removal in die casting molds.

[0010] The present invention proposes an adaptive control system for demolding and part removal of die-casting molds, comprising a die-casting machine, a sensing module, a control center, an execution module, and an emergency stop button installed on the surface of the die-casting machine.

[0011] The sensing module includes a vision sensor, a pressure sensor, a temperature sensor, a laser sensor, a vibration sensor, and an ultrasonic thickness sensor.

[0012] The control center uses an industrial computer running a real-time operating system, including the following functions:

[0013] A1. Data preprocessing.

[0014] A2. Based on the mold status: waiting to be painted, picking parts, or idle.

[0015] A3. Adaptive algorithms: including PID control and machine learning models.

[0016] A4. Data storage.

[0017] A5. Connected to the die-casting machine and the execution module via Ethernet.

[0018] The execution module performs specific operations according to the instructions of the control center, including a demolding device and a part-grabbing device. The demolding device includes a capillary nozzle for spraying a release agent, and the part-grabbing device includes a pneumatic silicone gripper for gripping the part.

[0019] Preferably, the vision sensor is mounted on the surface of the demolding device and is used to detect the position of the part; the pressure sensor is embedded inside the pneumatic silicone gripper and is used to monitor the contact force during part removal; the temperature sensor is mounted on the surface of the die casting machine and is used to detect the mold temperature; the laser sensor is mounted on the surface of the die casting machine and is used to monitor the mold position; the vibration sensor is mounted on the surface of the die casting machine and is used to monitor mold vibration; and the ultrasonic thickness sensor is mounted on the surface of the die casting machine and is used to detect the coating thickness on the mold surface.

[0020] Preferably, the adaptive algorithm of the control center includes:

[0021] The PID control algorithm for spraying pressure is as follows:

[0022]

[0023] in, For the controller's output signal, The difference between the set value and the actual measured value. For proportional gain, For integral gain, For differential gain, This is a dummy variable in integration operations;

[0024] Adaptive gain scheduling, the formula is as follows:

[0025]

[0026]

[0027]

[0028] in, , , For adaptive PID parameters, , , These are the baseline PID parameters, a set of optimal PID parameters carefully tuned under design conditions. , , For scheduling coefficients, As a scheduling variable, it can be the temperature difference between the current mold temperature monitored by the temperature sensor and the reference temperature, or the mold wear detected by the ultrasonic thickness sensor, which increases with the number of production cycles;

[0029] The learning model uses a random forest regression model, with the following formula:

[0030]

[0031] in, These are the model's predicted values, including optimal spraying time and optimal release agent dosage. For the prediction model function, The feature vector contains historical data on mold temperature, mold wear, and spraying pressure. This represents the total number of decision trees included in the random forest regression model. For the first Each decision tree has input features The prediction results;

[0032] The loss function is expressed in the following formula:

[0033]

[0034] in, For loss function, The number of samples in the training dataset, For the first The true value of each sample For the model to the first The predicted value for each sample, This is a regularization term used to prevent the model from overfitting. The regularization coefficient is . This is a penalty term for model complexity.

[0035] Preferably, the demolding device further includes a self-driven slide rail assembly installed on the upper surface of the die-casting machine. The surface of the self-driven slide rail assembly is provided with a slider, and a support plate is fixedly connected between two sliders. A hydraulic cylinder is fixedly installed on the upper surface of the support plate. One end of the piston rod of the hydraulic cylinder passes through the support plate and the upper surface of the die-casting machine in sequence and is fixedly connected to a connecting rod. A limit groove is opened on the upper surface of the die-casting machine. The surface of the piston rod of the hydraulic cylinder is slidably connected to the inner wall of the limit groove. The capillary nozzle is installed on the surface of the connecting rod.

[0036] Through the above technical solution, the extension and retraction of the hydraulic cylinder piston rod drives the connecting rod connected to it to move up and down. The up and down movement of the connecting rod drives the capillary nozzle to spray the mold release agent onto the mold surface. At the same time, in order to prevent affecting the normal movement trajectory of the mold, a self-driven slide rail is used to drive the slider to move, thereby driving the support plate to move. The movement of the support plate drives the hydraulic cylinder, connecting rod, and capillary nozzle away from the mold.

[0037] Preferably, the outer surface of the die-casting machine is provided with a tank, and a pump body is installed on the upper surface of the tank. The water suction end of the pump body extends into the interior of the tank through a pipe, and the water discharge end of the pump body extends into the interior of the connecting rod through a connecting pipe and is fixedly connected to one end of the capillary nozzle. A solenoid valve is installed on the surface of the connecting pipe.

[0038] By using the above technical solution, the solenoid valve is opened, and the pump body introduces the mold release agent in the tank into the capillary nozzle on the connecting rod through the connecting pipe, so that it is sprayed onto the mold surface through the capillary nozzle.

[0039] Preferably, a filter screen is installed below the mold of the die-casting machine, and a liquid collection tank is provided inside the die-casting machine. Both the liquid collection tank and the tank body are equipped with water level sensors.

[0040] Through the above technical solution, the filter screen filters the liquid generated after demolding, so that the filtered liquid falls into the liquid collection tank. The water level sensor in the tank monitors the water level in the tank. When the water level in the tank is lower than the preset threshold, an alarm is issued to remind the staff to add the release agent. The water level sensor in the liquid collection tank monitors the water level in the liquid collection tank. When the water level in the liquid collection tank is higher than the preset threshold, an alarm is issued to remind the staff to take action.

[0041] Preferably, a drive motor is fixedly mounted on one side surface of the die-casting machine, a groove is formed on the inner wall of the die-casting machine, a screw is mounted on the inner wall of the groove through a bearing, one end of the output shaft of the drive motor is fixedly sleeved with one end of the screw, a moving block is threadedly sleeved on the surface of the screw, the surface of the moving block is slidably engaged with the inner wall of the groove, guide rods are symmetrically distributed on the inner wall of the groove, the screw is located between two of the guide rods, the moving block is slidably sleeved with the surface of the guide rod, and the drive motor is electrically connected to the control center.

[0042] Through the above technical solution, the control center indirectly judges the accumulation of impurities on the surface of the filter screen based on the monitoring of the water level sensor in the liquid collection tank, and then issues start and stop commands. The rotation of the output shaft of the drive motor drives the screw connected to it to rotate, and the rotation of the screw drives the moving block to move along the surface of the guide rod and the inner wall of the groove.

[0043] Preferably, the moving block has sleeve rods arranged in a rectangular array on the side surface away from the drive motor. The surfaces of the three sleeve rods are slidably connected to the surfaces of the two guide rods and the screw, respectively. A scraper is fixedly connected to one end of the three sleeve rods extending out of the groove. The surface of the scraper is in slidable contact with the surface of the filter screen. A collection box is provided on the side surface of the die-casting machine near the filter screen.

[0044] Through the above technical solution, the movement of the moving block drives the three sleeve rods connected to it to move along the surfaces of the guide rod and the screw rod respectively. The movement of the three sleeve rods drives the scraper to clean the surface of the filter screen, and the cleaned debris falls into the collection box. At the same time, a pressure sensor is installed on the scraper. This pressure sensor is integrated with the pressure sensor in the sensing module to provide real-time feedback on the contact pressure between the scraper and the filter screen. The data is transmitted to the control center, which dynamically adjusts the speed of the drive motor to avoid damage to the filter screen.

[0045] Preferably, the part-retrieving device further includes an L-shaped support fixedly installed on the surface of the die-casting machine. A push cylinder is embedded in the surface of the L-shaped support. A push plate is fixedly connected to one end of the piston rod of the push cylinder. A rotary cylinder is fixedly installed on the upper surface of the push plate. A mounting housing is fixedly connected to the clamping arm of the rotary cylinder. A rotary motor is fixedly installed inside the mounting housing. A connecting shaft is fixedly sleeved at one end of the output shaft of the rotary motor through a coupling. The pneumatic silicone gripper is installed at the end of the connecting shaft away from the mounting housing.

[0046] Through the above technical solution, the extension and retraction of the cylinder piston rod drives the push plate connected to it to move. The movement of the push plate drives the rotary cylinder to move, and in conjunction with the rotary cylinder, it facilitates the rotation of the mounting housing, so that the pneumatic silicone gripper approaches the mold. The rotation of the output shaft of the rotary motor drives the connecting shaft to rotate 90°, so that the pneumatic silicone gripper rotates 90°, thereby facilitating the clamping of parts on the mold and the placement of the parts.

[0047] The present invention proposes a method for using an adaptive control system for demolding and part removal in die casting molds, comprising the following steps:

[0048] S1. Start the system. The control center performs self-checks on each module, including sensor calibration of the sensing module, and zeroing of the hydraulic cylinder, push cylinder, rotary cylinder, and rotary motor of the execution module; load the stored adaptive parameters and machine learning models.

[0049] S2. The sensing module monitors the operating status of the die-casting machine in real time, including mold temperature, mold position, vibration, and detects the end signal of the die-casting cycle.

[0050] S3. The vision sensor collects images of the mold surface, and the control center analyzes the degree of residue through image processing. Based on the analysis results and the mold wear data detected by the ultrasonic thickness sensor, the optimal spraying pressure, release agent dosage, and demolding time are predicted by combining the machine learning model.

[0051] S4. When demolding is required, start the self-driven slide rail assembly, adjust the position of the slider and support plate so that the capillary nozzle is aligned with the mold, the hydraulic cylinder piston rod extends, driving the connecting rod to approach the mold surface, the solenoid valve and pump body are opened, the release agent is drawn from the tank and sprayed onto the mold surface through the capillary nozzle, the pressure sensor and temperature sensor provide real-time feedback data, the control center receives the real-time data from the pressure sensor and temperature sensor, and dynamically adjusts the spraying pressure and time through PID control.

[0052] S5. During the spraying process, the waste liquid flows into the collection tank after being filtered by the filter screen. The water level sensor monitors the water level in the tank and the collection tank. When the water level exceeds the threshold, an alarm is issued. At the same time, the control center indirectly judges the accumulation of impurities on the surface of the filter screen based on the monitoring of the water level sensor in the collection tank. Then, it issues a command to start the drive motor, which drives the screw to rotate, so that the moving block and scraper clean the impurities on the surface of the filter screen. The impurities fall into the collection tank.

[0053] S6. When it is necessary to remove the part, the push cylinder pushes the push plate and the rotary cylinder to move, so that the pneumatic silicone gripper is close to the mold. The rotary cylinder adjusts the position of the gripper. The rotary motor drives the connecting shaft and the gripper to rotate to the gripping position. The pneumatic silicone gripper grips the part. The pressure sensor monitors the clamping force. The rotary cylinder and the push cylinder work together to move the part out and place it in the designated position.

[0054] S7. The control center records the operation data and sensor data of each demolding and part removal, updates the machine learning model, optimizes the PID parameters and demolding strategy based on historical data, and achieves adaptive control.

[0055] S8, vibration sensor and temperature sensor continuously monitor the equipment status. If abnormal vibration or temperature exceeds the limit, the control center will immediately stop operation and alarm. The emergency stop button can be manually triggered at any stage to immediately cut off the power to the execution module and enter a safe state.

[0056] The beneficial effects of this invention are as follows:

[0057] 1. Through adaptive PID algorithm and machine learning model, the system can dynamically adjust the spraying pressure, time and release agent dosage of the demolding device according to the mold temperature, residue on the mold and mold wear monitored in real time by the sensing module, so as to ensure the best demolding effect, save resources and avoid mold damage.

[0058] 2. By setting up a demolding device, and utilizing a filter screen, liquid collection tank, and scraper, the system can automatically filter, recover waste liquid, and clean the filter screen, which facilitates the continuous and stable operation of the system, reduces maintenance frequency, and improves environmental friendliness. Attached Figure Description

[0059] Figure 1 This is a schematic diagram of an adaptive control system and method for demolding and removing parts from a die-casting mold, as proposed in this invention.

[0060] Figure 2 This is a three-dimensional view of the filter screen structure of an adaptive control system and method for demolding and removing parts for a die-casting mold proposed in this invention.

[0061] Figure 3 This is a perspective view of the scraper structure of an adaptive control system and method for demolding and removing parts for a die-casting mold proposed in this invention.

[0062] Figure 4 This is a three-dimensional view of the water level sensor structure of the adaptive control system and method for demolding and part removal of die-casting molds proposed in this invention;

[0063] Figure 5 This is a three-dimensional view of the soft nylon brush structure of the adaptive control system and method for demolding and removing parts for die casting molds proposed in this invention;

[0064] Figure 6 This is a three-dimensional view of the capillary nozzle structure of an adaptive control system and method for demolding and removing parts for die casting molds proposed in this invention.

[0065] Figure 7 This is a perspective view of the L-shaped support structure of the adaptive control system and method for demolding and removing parts for die casting molds proposed in this invention.

[0066] Figure 8 This is a three-dimensional view of the rotary motor structure of the adaptive control system and method for demolding and removing parts for die casting molds proposed in this invention.

[0067] Figure 9 This is a system architecture block diagram of an adaptive control system and method for demolding and removing parts for die casting molds proposed in this invention;

[0068] Figure 10 This is a system flowchart of an adaptive control system and method for demolding and removing parts for die casting molds proposed in this invention.

[0069] Figure 11 This is a flowchart of the adaptive control algorithm for an adaptive control system and method for demolding and removing parts from a die-casting mold, as proposed in this invention.

[0070] In the diagram: 1. Die-casting machine; 2. Emergency stop button; 3. Capillary nozzle; 31. Self-driven slide rail assembly; 32. Slider; 33. Support plate; 34. Hydraulic cylinder; 35. Connecting rod; 36. Limiting groove; 37. Water tank; 38. Water pump; 39. Connecting pipe; 310. Solenoid valve; 311. Filter screen; 312. Liquid collection tank; 313. Water level sensor; 314. Drive motor; 315. Groove; 316. Screw; 317. Moving block; 318. Guide rod; 319. Sleeve rod; 320. Scraper; 321. Collection box; 4. Pneumatic silicone gripper; 41. L-shaped support; 42. Push cylinder; 43. Push plate; 44. Rotary cylinder; 45. Mounting housing; 46. Rotary motor; 47. Connecting shaft. Detailed Implementation

[0071] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments.

[0072] Reference Figures 1-11 An adaptive control system for demolding and part removal of die-casting molds includes a die-casting machine 1, a sensing module, a control center, an execution module, and an emergency stop button 2 mounted on the surface of the die-casting machine 1.

[0073] The sensing module includes a vision sensor, a pressure sensor, a temperature sensor, a laser sensor, a vibration sensor, and an ultrasonic thickness sensor.

[0074] To monitor the mold condition in real time, a vision sensor is installed on the surface of the demolding device to detect the position of the part; a pressure sensor is embedded inside the pneumatic silicone gripper 4 to monitor the contact force during part removal; a temperature sensor is installed on the surface of the die-casting machine 1 to detect the mold temperature; a laser sensor is installed on the surface of the die-casting machine 1 to monitor the mold position; a vibration sensor is installed on the surface of the die-casting machine 1 to monitor mold vibration; and an ultrasonic thickness sensor is installed on the surface of the die-casting machine 1 to detect the coating thickness on the mold surface.

[0075] like Figures 9-11 As shown, the control center uses an industrial PC running a real-time operating system, which includes the following functions:

[0076] A1. Data preprocessing.

[0077] A2. Based on the mold status: waiting to be painted, picking parts, or idle.

[0078] A3. Adaptive algorithms: including PID control and machine learning models.

[0079] A4. Data storage.

[0080] A5. Connect to the die-casting machine 1 and execution module via Ethernet.

[0081] The adaptive algorithm of the control center includes:

[0082] The PID control algorithm for spraying pressure is as follows:

[0083]

[0084] in, For the controller's output signal, The difference between the set value and the actual measured value. For proportional gain, For integral gain, For differential gain, This is a dummy variable in integration operations;

[0085] Adaptive gain scheduling, the formula is as follows:

[0086]

[0087]

[0088]

[0089] in, , , For adaptive PID parameters, , , These are the baseline PID parameters, a set of optimal PID parameters carefully tuned under design conditions. , , For scheduling coefficients, As a scheduling variable, it can be the temperature difference between the current mold temperature monitored by the temperature sensor and the reference temperature, or the mold wear detected by the ultrasonic thickness sensor, which increases with the number of production cycles;

[0090] The learning model uses a random forest regression model, with the following formula:

[0091]

[0092] in, These are the model's predicted values, including optimal spraying time and optimal release agent dosage. For the prediction model function, The feature vector contains historical data on mold temperature, mold wear, and spraying pressure. This represents the total number of decision trees included in the random forest regression model. For the first Each decision tree has input features The prediction results;

[0093] The loss function is expressed in the following formula:

[0094]

[0095] in, For loss function, The number of samples in the training dataset, For the first The true value of each sample For the model to the first The predicted value for each sample, This is a regularization term used to prevent the model from overfitting. The regularization coefficient is . This is a penalty term for model complexity.

[0096] like Figures 1-8 As shown, the execution module performs specific operations according to the instructions of the control center, including a demolding device and a part-grabbing device. The demolding device includes a capillary nozzle 3 for spraying demolding agent, and the part-grabbing device includes a pneumatic silicone gripper 4 for gripping parts.

[0097] like Figures 1-6 As shown, to facilitate demolding, the demolding device also includes a self-driven slide rail assembly 31 installed on the upper surface of the die-casting machine 1. The surface of the self-driven slide rail assembly 31 is provided with a slider 32, and a support plate 33 is fixedly connected between two sliders 32. A hydraulic cylinder 34 is fixedly installed on the upper surface of the support plate 33. One end of the piston rod of the hydraulic cylinder 34 passes through the support plate 33 and the upper surface of the die-casting machine 1 in sequence and is fixedly connected to a connecting rod 35. A limit groove 36 is opened on the upper surface of the die-casting machine 1. The surface of the piston rod of the hydraulic cylinder 34 is slidably connected to the inner wall of the limit groove 36. The capillary nozzle 3 is installed on the surface of the connecting rod 35. The extension and retraction of the piston rod of the hydraulic cylinder 34 drives the connecting rod 35 connected to it to move up and down. At the same time, in order to prevent affecting the normal movement trajectory of the mold, the self-driven slide rail assembly 31 drives the slider 32 to move, thereby driving the support plate 33 to move. The movement of the support plate 33 drives the hydraulic cylinder 34, the connecting rod 35, and the capillary nozzle 3 away from the mold.

[0098] In order to spray the release agent onto the mold surface, a tank 37 is provided on the outer surface of the die casting machine 1. A pump body 38 is installed on the upper surface of the tank 37. The suction end of the pump body 38 extends into the interior of the tank 37 through a pipe. The outlet end of the pump body 38 extends into the interior of the connecting rod 35 through a connecting pipe 39 and is fixedly connected to one end of the capillary nozzle 3. A solenoid valve 310 is installed on the surface of the connecting pipe 39. When the solenoid valve 310 is opened, the pump body 38 introduces the release agent in the tank 37 into the capillary nozzle 3 on the connecting rod 35 through the connecting pipe 39, so that it is sprayed onto the mold surface through the capillary nozzle 3.

[0099] To filter the liquid, a filter screen 311 is installed below the mold of the die-casting machine 1. A liquid collection tank 312 is installed inside the die-casting machine 1. Water level sensors 313 are installed inside both the liquid collection tank 312 and the tank body 37. The filter screen 311 filters the liquid generated after demolding, so that the filtered liquid falls into the liquid collection tank 312. The water level sensor 313 in the tank body 37 monitors the water level in the tank body 37. When the water level in the tank body 37 is lower than a preset threshold, an alarm is issued to remind the staff to add release agent. The water level sensor 313 in the liquid collection tank 312 monitors the water level in the liquid collection tank 312. When the water level in the liquid collection tank 312 is higher than a preset threshold, an alarm is issued to remind the staff to take action.

[0100] To clean the filter screen 311, a drive motor 314 is fixedly installed on one side of the die-casting machine 1. A groove 315 is formed on the inner wall of the die-casting machine 1. A screw 316 is mounted on the inner wall of the groove 315 via bearings. One end of the output shaft of the drive motor 314 is fixedly sleeved with one end of the screw 316. A moving block 317 is threaded onto the surface of the screw 316. The surface of the moving block 317 slides against the inner wall of the groove 315. Guide rods 318 are symmetrically distributed on the inner wall of the groove 315. Located between two guide rods 318, the moving block 317 is slidably sleeved with the surface of the guide rods 318. The drive motor 314 is electrically connected to the control center. The control center indirectly judges the accumulation of impurities on the surface of the filter screen 311 based on the monitoring of the water level sensor 313 in the liquid collection tank 312, and then issues start and stop commands. The rotation of the output shaft of the drive motor 314 drives the screw 316 connected to it to rotate. The rotation of the screw 316 drives the moving block 317 to move along the surface of the guide rod 318 and the inner wall of the groove 315.

[0101] On the side of the moving block 317 away from the drive motor 314, three sleeve rods 319 are installed in a rectangular array. The surfaces of the three sleeve rods 319 are slidably connected to the surfaces of the two guide rods 318 and the screw 316, respectively. A scraper 320 is fixedly connected to one end of the three sleeve rods 319 extending out of the groove 315. The surface of the scraper 320 is in slidable contact with the surface of the filter screen 311. A collection box 321 is provided on the side of the die-casting machine 1 near the filter screen 311. The movement of the moving block 317 drives the three sleeve rods 319 connected to it. The three sleeve rods 319 move along the surfaces of the guide rod 318 and the screw 316 respectively. The movement of the three sleeve rods 319 drives the scraper 320 to clean the surface of the filter screen 311, and the cleaned debris falls into the collection box 321. At the same time, a pressure sensor is installed on the scraper 320. This pressure sensor is integrated with the pressure sensor in the sensing module and provides real-time feedback on the contact pressure between the scraper 320 and the filter screen 311. The data is transmitted to the control center, which dynamically adjusts the speed of the drive motor 314 to prevent damage to the filter screen 311.

[0102] By setting up a demolding device, the system utilizes a filter screen 311, a liquid collection tank 312, and a scraper 320 to enable automatic filtration, waste liquid recovery, and self-cleaning of the filter screen 311. This facilitates the continuous and stable operation of the system, reduces maintenance frequency, and improves environmental friendliness.

[0103] like Figure 1 and Figures 7-8 As shown, in order to remove the parts from the mold, the part removal device also includes an L-shaped support 41 fixedly mounted on the surface of the die-casting machine 1. A push cylinder 42 is embedded in the surface of the L-shaped support 41. A push plate 43 is fixedly connected to one end of the piston rod of the push cylinder 42. A rotary cylinder 44 is fixedly mounted on the upper surface of the push plate 43. A mounting housing 45 is fixedly connected to the clamping arm of the rotary cylinder 44. A rotary motor 46 is fixedly mounted inside the mounting housing 45. A connecting shaft 47 is fixedly sleeved to one end of the output shaft of the rotary motor 46 through a coupling. The pneumatic silicone gripper 4 is installed at the end of the connecting shaft 47 away from the mounting housing 45. By pushing the piston rod of the cylinder 42 to extend or retract, it drives the push plate 43 connected to it to move. The movement of the push plate 43 drives the rotary cylinder 44 to move. In conjunction with the rotary cylinder 44, it facilitates the rotation of the mounting housing 45, so that the pneumatic silicone gripper 4 is close to the mold. The rotation of the output shaft of the rotary motor 46 drives the connecting shaft 47 to rotate 90°, so that the pneumatic silicone gripper 4 rotates 90°, thereby facilitating the clamping of parts on the mold and the placement of the parts.

[0104] Through adaptive PID algorithm and machine learning model, the system can dynamically adjust the spraying pressure, time and release agent dosage of the demolding device based on the mold temperature, residue on the mold and mold wear monitored in real time by the sensing module, to ensure the best demolding effect, while also saving resources and avoiding mold damage.

[0105] refer to Figures 1-11 A method for using an adaptive control system for demolding and part removal in die casting molds includes the following steps:

[0106] S1. Start the system. The control center performs self-checks on each module, including sensor calibration of the sensing module, and zeroing of the hydraulic cylinder 34, push cylinder 42, rotary cylinder 44, and rotary motor 46 of the execution module; load the stored adaptive parameters and machine learning model.

[0107] S2. The sensing module monitors the operating status of the die-casting machine 1 in real time, including mold temperature, mold position, vibration, and detects the end signal of the die-casting cycle.

[0108] S3. The vision sensor collects images of the mold surface, and the control center analyzes the degree of residue through image processing. Based on the analysis results and the mold wear data detected by the ultrasonic thickness sensor, the optimal spraying pressure, release agent dosage, and demolding time are predicted by combining the machine learning model.

[0109] S4. When demolding is required, start the self-driven slide rail assembly 31, adjust the position of the slider 32 and the support plate 33 so that the capillary nozzle 3 is aligned with the mold, the piston rod of the hydraulic cylinder 34 extends, driving the connecting rod 35 to approach the mold surface, and open the solenoid valve 310 and the pump body 38 to extract the release agent from the tank 37 and spray it onto the mold surface through the capillary nozzle 3. The pressure sensor and temperature sensor provide real-time feedback data. The control center receives the real-time data from the pressure sensor and temperature sensor and dynamically adjusts the spraying pressure and time through PID control.

[0110] S5. During the spraying process, the waste liquid flows into the collection tank 312 after being filtered by the filter screen 311. The water level sensor 313 monitors the water level in the tank 37 and the collection tank 312. When the water level exceeds the threshold, an alarm is issued. At the same time, the control center indirectly judges the accumulation of impurities on the surface of the filter screen 311 based on the monitoring of the water level sensor 313 in the collection tank 312. Then, it issues a command to start the drive motor 314, which drives the screw 316 to rotate, so that the moving block 317 and the scraper 320 clean the impurities on the surface of the filter screen 311. The impurities fall into the collection box 321.

[0111] S6. When it is necessary to remove the part, the push cylinder 42 pushes the push plate 43 and the rotary cylinder 44 to move, so that the pneumatic silicone gripper 4 approaches the mold. The rotary cylinder 44 adjusts the position of the gripper. The rotary motor 46 drives the connecting shaft 47 and the gripper to rotate to the gripping position. The pneumatic silicone gripper 4 grips the part. The pressure sensor monitors the clamping force. The rotary cylinder 44 and the push cylinder 42 work together to move the part out and place it in the designated position.

[0112] S7. The control center records the operation data and sensor data of each demolding and part removal, updates the machine learning model, optimizes the PID parameters and demolding strategy based on historical data, and achieves adaptive control.

[0113] S8, vibration sensor and temperature sensor continuously monitor the equipment status. If abnormal vibration or temperature exceeds the limit, the control center will immediately stop operation and alarm. Emergency stop button 2 can be manually triggered at any stage to immediately cut off the power to the execution module and enter a safe state.

[0114] The above description is only a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any equivalent substitutions or modifications made by those skilled in the art within the scope of the technology disclosed in the present invention, based on the technical solution and inventive concept of the present invention, should be covered within the scope of protection of the present invention.

Claims

1. An adaptive control system for demolding and part removal in die casting molds, characterized in that: Includes a die-casting machine (1), a sensing module, a control center, an execution module, and an emergency stop button (2) installed on the surface of the die-casting machine (1); The sensing module includes a visual sensor, a pressure sensor, a temperature sensor, a laser sensor, a vibration sensor, and an ultrasonic thickness sensor. The control center uses an industrial computer running a real-time operating system, including the following functions: A1. Data preprocessing; A2. Based on the mold status: waiting to be painted, in the process of picking parts, or idle mode; A3. Adaptive algorithms: including PID control and machine learning models; A4. Data storage; A5. Connected to the die-casting machine (1) and the execution module via Ethernet; The execution module performs specific operations according to the instructions of the control center, including a demolding device and a part-grabbing device. The demolding device includes a capillary nozzle (3) for spraying demolding agent, and the part-grabbing device includes a pneumatic silicone gripper (4) for gripping parts. The adaptive algorithm of the control center includes: The PID control algorithm for spraying pressure is as follows: ； in, For the controller's output signal, The difference between the set value and the actual measured value. For proportional gain, For integral gain, For differential gain, This is a dummy variable in integration operations; Adaptive gain scheduling, the formula is as follows: ； ； ； in, , , For adaptive PID parameters, , , These are the baseline PID parameters, a set of optimal PID parameters carefully tuned under design conditions. , , For scheduling coefficients, As a scheduling variable, it can be the temperature difference between the current mold temperature monitored by the temperature sensor and the reference temperature, and the mold wear detected by the ultrasonic thickness sensor, which increases with the number of production cycles; The learning model uses a random forest regression model, with the following formula: ； in, These are the model's predicted values, including optimal spraying time and optimal release agent dosage. For the prediction model function, The feature vector contains historical data on mold temperature, residue adhesion, mold wear, and spraying pressure. This represents the total number of decision trees included in the random forest regression model. For the first Each decision tree has input features The prediction results; The loss function is expressed in the following formula: ； in, For loss function, The number of samples in the training dataset, For the first The true value of each sample For the model to the first The predicted value for each sample, This is a regularization term used to prevent the model from overfitting. The regularization coefficient is . This is a penalty term for model complexity.

2. The adaptive control system for demolding and part removal of a die-casting mold according to claim 1, characterized in that: The vision sensor is installed on the surface of the demolding device and is used to detect the position of the part. The pressure sensor is embedded inside the pneumatic silicone gripper (4) and is used to monitor the contact force when picking up the part. The temperature sensor is installed on the surface of the die casting machine (1) and is used to detect the mold temperature. The laser sensor is installed on the surface of the die casting machine (1) and is used to monitor the mold position. The vibration sensor is installed on the surface of the die casting machine (1) and is used to monitor the mold vibration. The ultrasonic thickness sensor is installed on the surface of the die casting machine (1) and is used to detect the coating thickness on the mold surface.

3. The adaptive control system for demolding and part removal of a die-casting mold according to claim 1, characterized in that: The demolding device also includes a self-driven slide rail assembly (31) installed on the upper surface of the die casting machine (1). The surface of the self-driven slide rail assembly (31) is provided with a slider (32). A support plate (33) is fixedly connected between the two sliders (32). A hydraulic cylinder (34) is fixedly installed on the upper surface of the support plate (33). One end of the piston rod of the hydraulic cylinder (34) passes through the support plate (33) and the upper surface of the die casting machine (1) in sequence and is fixedly connected to a connecting rod (35). A limit groove (36) is opened on the upper surface of the die casting machine (1). The surface of the piston rod of the hydraulic cylinder (34) is slidably connected to the inner wall of the limit groove (36). The capillary nozzle (3) is installed on the surface of the connecting rod (35).

4. The adaptive control system for demolding and part removal of a die-casting mold according to claim 3, characterized in that: The outer surface of the die-casting machine (1) is provided with a tank (37), and a pump body (38) is installed on the upper surface of the tank (37). The suction end of the pump body (38) extends into the interior of the tank (37) through a pipe, and the outlet end of the pump body (38) extends into the interior of the connecting rod (35) through a connecting pipe (39) and is fixedly connected to one end of the capillary nozzle (3). A solenoid valve (310) is installed on the surface of the connecting pipe (39).

5. The adaptive control system for demolding and part removal of a die-casting mold according to claim 4, characterized in that: A filter screen (311) is installed below the mold of the die casting machine (1). A liquid collection tank (312) is provided inside the die casting machine (1). A water level sensor (313) is installed inside both the liquid collection tank (312) and the tank body (37).

6. The adaptive control system for demolding and part removal of a die-casting mold according to claim 5, characterized in that: A drive motor (314) is fixedly installed on one side surface of the die casting machine (1). A groove (315) is provided on the inner wall of the die casting machine (1). A screw (316) is installed on the inner wall of the groove (315) through a bearing. One end of the output shaft of the drive motor (314) is fixedly sleeved with one end of the screw (316). A moving block (317) is threadedly sleeved on the surface of the screw (316). The surface of the moving block (317) is slidably engaged with the inner wall of the groove (315). Guide rods (318) are symmetrically distributed on the inner wall of the groove (315). The screw (316) is located between the two guide rods (318). The moving block (317) is slidably sleeved with the surface of the guide rod (318). The drive motor (314) is electrically connected to the control center.

7. The adaptive control system for demolding and part removal of a die-casting mold according to claim 6, characterized in that: The moving block (317) has sleeve rods (319) arranged in a rectangular array on the side surface away from the drive motor (314). The surfaces of the three sleeve rods (319) are slidably connected to the surfaces of the two guide rods (318) and the screw (316), respectively. A scraper (320) is fixedly connected to one end of the three sleeve rods (319) extending out of the groove (315). The surface of the scraper (320) is in slidable contact with the surface of the filter screen (311). A collection box (321) is provided on the side surface of the die-casting machine (1) near the filter screen (311).

8. The adaptive control system for demolding and part removal of a die-casting mold according to claim 1, characterized in that: The part-removing device also includes an L-shaped support (41) fixedly installed on the surface of the die-casting machine (1). A push cylinder (42) is embedded on the surface of the L-shaped support (41). A push plate (43) is fixedly connected to one end of the piston rod of the push cylinder (42). A rotary cylinder (44) is fixedly installed on the upper surface of the push plate (43). A mounting housing (45) is fixedly connected to the clamping arm of the rotary cylinder (44). A rotary motor (46) is fixedly installed inside the mounting housing (45). A connecting shaft (47) is fixedly sleeved on one end of the output shaft of the rotary motor (46) through a coupling. The pneumatic silicone gripper (4) is installed on the end of the connecting shaft (47) away from the mounting housing (45).

9. A method of using the adaptive control system for demolding and part removal of a die-casting mold according to any one of claims 1-8, characterized in that, Includes the following steps: S1. Start the system and the control center performs self-checks on each module, including sensor calibration of the sensing module, and zeroing of the hydraulic cylinder (34), push cylinder (42), rotary cylinder (44), and rotary motor (46) of the execution module; load the stored adaptive parameters and machine learning model; S2. The sensing module monitors the operating status of the die casting machine (1) in real time, including mold temperature, mold position, vibration, and detects the end signal of the die casting cycle. S3. The vision sensor collects images of the mold surface, and the control center analyzes the degree of residue through image processing. Based on the analysis results and the mold wear data detected by the ultrasonic thickness sensor, the optimal spraying pressure, release agent dosage, and demolding time are predicted by combining the machine learning model. S4. When demolding is required, start the self-driven slide rail group (31), adjust the position of the slider (32) and the support plate (33) so that the capillary nozzle (3) is aligned with the mold. The piston rod of the hydraulic cylinder (34) extends and drives the connecting rod (35) to approach the mold surface. The solenoid valve (310) and the pump body (38) are opened to extract the demolding agent from the tank (37) and spray it onto the mold surface through the capillary nozzle (3). The pressure sensor and temperature sensor provide real-time feedback data. The control center receives the real-time data from the pressure sensor and temperature sensor and dynamically adjusts the spraying pressure and time through PID control. S5. During the spraying process, the waste liquid flows into the collection tank (312) after being filtered by the filter screen (311). The water level sensor (313) monitors the water level of the tank (37) and the collection tank (312). When the water level exceeds the threshold, an alarm is issued. At the same time, the control center indirectly judges the accumulation of impurities on the surface of the filter screen (311) based on the monitoring of the water level sensor (313) in the collection tank (312). Thus, it issues a command to start the drive motor (314), which drives the screw (316) to rotate, so that the moving block (317) and the scraper (320) clean the impurities on the surface of the filter screen (311). The impurities fall into the collection box (321). S6. When it is necessary to remove the part, push cylinder (42) pushes push plate (43) and rotary cylinder (44) to move, so that pneumatic silicone gripper (4) approaches the mold. Rotary cylinder (44) adjusts the gripper position. Rotary motor (46) drives connecting shaft (47) and gripper to rotate to gripping position. Pneumatic silicone gripper (4) grips part. Pressure sensor monitors clamping force. Rotary cylinder (44) and push cylinder (42) cooperate to remove part and place it in designated position. S7. The control center records the operation data and sensor data of each demolding and part removal, updates the machine learning model, optimizes the PID parameters and demolding strategy based on historical data, and achieves adaptive control. S8, vibration sensor and temperature sensor continuously monitor the equipment status. If abnormal vibration or temperature exceeds the limit, the control center will immediately stop operation and alarm. The emergency stop button (2) can be manually triggered at any stage to immediately cut off the power supply of the execution module and enter a safe state.

Images