Tablet press mold production control method and production line thereof

Abstract

The invention relates to the technical field of mold production and manufacturing, in particular to a tablet press mold production control method and a production line thereof.A closed-loop production system of AGV intelligent logistics, a manipulator and a central control system is constructed, an AGV automatically recognizes tray codes and plans a path according to an MES instruction, the collision risk caused by manual carrying is eliminated, and the production efficiency is improved. And the transfer efficiency is improved by 60% or above, all machining devices are linked through distributed manipulators, unmanned operation of the whole process of machining, detection and transfer is achieved, and the labor cost is reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of mold production and manufacturing, and in particular to a production control method and production line for tablet press molds. Background Art

[0002] At present, tablet press molds are the core components used for pressing and forming sheet products in industries such as pharmaceuticals, food, and chemicals. The production of tablet press molds in the existing technology has the following technical problems: 1. In the production process of tablet press molds at home and abroad, each step has been produced by a single machine. After completing a single step and passing the inspection, it is manually transferred to the next step. Due to the large number of production steps of tablet press molds, a large amount of manpower and time are consumed in the inspection and transfer process. During the transfer process, it is also difficult to avoid quality problems caused by bumps of workpieces, etc. 2. Isolated processes lead to error accumulation: Each process of the traditional production line runs independently, lacking real-time data interaction. The processing errors of the previous process cannot be transmitted to the subsequent process for compensation, resulting in the amplification of errors step by step. For example, the size deviation in rough turning directly affects the qualified rate of the fine grinding process, and it is necessary to rely on manual re-inspection to intercept defective parts.

[0003] 3. Quality inspection is lagging and one-sided: Relying on manual sampling inspection and single-point measurement, it is impossible to achieve full-area precision scanning of the processing surface. The abnormal discovery of key dimensions (such as cavity flatness and assembly clearance) is seriously lagging, and the defect positioning accuracy is low, resulting in low rework efficiency.

[0004] Therefore, a production control method and production line for tablet press molds that can solve the above problems are needed to improve the production efficiency and product quality of tablet press molds. Summary of the Invention

[0005] The present invention provides a production control method and production line for tablet press molds. By constructing a closed-loop production system of "AGV intelligent logistics + manipulator + central control system", the AGV vehicle automatically identifies the tray code and plans the path according to the MES instruction, eliminating the risk of bumps caused by manual handling, and the transfer efficiency is increased by more than 60%. Each processing equipment is linked by a distributed manipulator to realize the unmanned operation of the whole process of "processing - inspection - transfer", reducing the labor cost.

[0006] The technical solution adopted by the present invention to solve the above technical problems is: A production control method for tablet press molds, comprising the following steps: Step 1: Saw the workpiece raw material into blank materials by a circular sawing machine to complete the stock preparation, put it into a tray, and the AGV vehicle goes to the tray to transport the workpiece to the first double-spindle CNC lathe for rough turning processing; Step 2: After rough turning the workpiece, perform the operation of surface grinding the end face, detect its flatness using a contact probe and compensate the grinding allowance in real time; Step 3: Perform profiling on the workpiece. During the profiling process, spiral scan the surface of the workpiece cavity using a multi-angle vision camera. If there are crack or deformation defects on the workpiece, automatically trigger the rework process; Step 4: Perform secondary rough turning and heat treatment on the profiled workpiece in sequence; Step 5: Perform polishing on the cavity of the workpiece after heat treatment, and then perform rough grinding on the polished workpiece through a first external cylindrical grinding machine and centerless grinding through a first centerless grinding machine in sequence; Step 6: Surface grind the working length and total length of the workpiece after centerless grinding; Step 7: Perform finish turning on the outer shape of the surface-ground workpiece. Before finish turning the outer shape, generate a differential machining path through visual scanning, and synchronously finish machining the head diameter and punch tail with a third double-spindle lathe; Step 8: Perform finish grinding on the small end of the workpiece after finish turning the outer shape through a second external cylindrical grinding machine and perform finish grinding on the punch body through a second centerless grinding machine in sequence; Establish a diameter data linkage mechanism between the processes of finish grinding the small end and finish grinding the punch body, and dynamically adjust the grinding feed rate; When finish grinding the small end, calibrate the grinding reference based on the total length data of the previous surface grinding, and verify the roundness error after finish grinding; During the process of finish grinding the punch body, real-time monitor the balance state of the grinding wheel through a vibration sensor, and dynamically adjust the grinding parameters in combination with the data of the on-line roundness meter.

[0007] Furthermore, the working process of the AGV vehicle includes the following steps: a. Receive instructions from the control system; b. Plan the route, identify the unique identification code of the pallet, locate the position of the pallet, and go to the specified pallet position; c. Grab the workpiece, avoid obstacles, and transport the workpiece along the preset path; d. After reaching the target position, place the workpiece on the specified tooling fixture or machine tool and confirm that the workpiece has been correctly placed; e. Send a confirmation signal to the control system, wait to receive the next instruction from the control system, and go back to step a to continue execution.

[0008] Furthermore, the AGV vehicle is configured with a radio frequency identification module. Before loading the workpiece, perform two-way verification with the process equipment by scanning the unique identification code of the pallet. When the identification code does not match the process instruction requirements of the task currently to be executed by the AGV vehicle, trigger an alarm and upload the exception code to the central control system.

[0009] Further, in the rough turning, secondary rough turning and finish turning processes, a line array camera is used to perform a 360° circumferential scan on the outer contour of the workpiece, and a dimensional error heat map is generated through point cloud comparison; In the flat grinding end face process, a contact probe scans the end face in a spiral path, collects flatness data of at least 50 sampling points, and calculates the grinding allowance distribution based on the least squares method to fit the plane equation; The collected data is uploaded to the MES system in real time and automatically compared with the process tolerance zone. When three consecutive workpieces are out of tolerance, the equipment self-check program is triggered.

[0010] Further, in the polishing process of step 5, the manipulator identifies the unique identification code of the tray, grabs the punch rod, and polishes the small head cavity. The following control method is adopted for the manipulator: a. The normal pressure and tangential friction force on the contact surface between the grinding head and the cavity are collected in real time through a six-axis force sensor; b. The pressure signal is converted into a polishing depth compensation amount, and the feed rate of the Z-axis servo motor is dynamically adjusted; c. According to the friction force threshold, a gradient adjustment rule for the polishing speed is set, and an emergency stop is triggered when the friction force exceeds the safety threshold.

[0011] Further, the generation method of the visual scanning difference machining path in step 7 includes: a. Obtain the three-dimensional point cloud data of the workpiece through a binocular stereo camera and fit and align it with the CAD model; b. Generate equally spaced cross-sectional contours along the axis, and calculate the radial deviation Δr between the actual contour and the theoretical contour of each cross-section; c. Fit the abnormal section where Δr > 0.1 mm with a B-spline curve to generate a tool compensation path; d. Decompose the tool compensation path into the main / auxiliary axis machining intervals of the double-spindle lathe, and set an overlapping cutting zone of 0.05 mm to eliminate the tool mark at the joint.

[0012] Further, the diameter data linkage mechanism in step 8 includes: in the finish grinding of the small head process, the small head diameter D1 and roundness δ1 are collected by the on-line roundness meter of the cylindrical grinder and written into the RFID tag; when the workpiece enters the finish grinding of the punch body process, the centerless grinder reads the small head diameter D1 and roundness δ1, and calculates the target diameter D2 of the punch body section according to the formula D2 = D1 - Δ, where Δ is the process reserved assembly gap, and adjusts the fluctuation coefficient K of the grinding wheel feed speed based on the roundness δ1 value, K = 1 + 0.2 × , where the upper tolerance limit (i.e., the maximum allowable tolerance of δ1) is 0.003 mm.

[0013] Further, the rework process in step 3 includes: Step 3-1: After the multi-angle vision camera detects a crack or deformation defect, it sends a defect code to the central control system; Step 3-2: The AGV vehicle receives the instruction and directly sends the defective workpiece from the molding station back to the starting end of the molding process; Step 3-3: If the defect type is a crack, the molding pressure automatically increases by 10% and the holding time is extended by 5 seconds; if the defect type is deformation, a local compensation pressure is increased at the die position corresponding to the defective area; Step 3-4: After molding, only the defective area of the workpiece is locally scanned, and the qualified judgment is completed within 3 seconds. If it is qualified, the workpiece enters the next process for secondary rough turning and heat treatment processing. If it is still unqualified, the workpiece is transferred to the waste area by the AGV vehicle.

[0014] A production line of a tablet press mold production control method has a first area, a second area, and a third area arranged in parallel. The first area is sequentially provided with a circular sawing machine, a first double-spindle CNC lathe, a first surface grinder, a column-type hydraulic press, and a second double-spindle CNC lathe; the second area is sequentially provided with a first external grinder, a first centerless grinder, a second surface grinder, and a third surface grinder; the third area is sequentially provided with a third double-spindle CNC lathe, a second external grinder, and a second centerless grinder; the heat treatment device is adjacent to the robotic arm polishing group and is arranged outside the first area, the second area, and the third area; among them, the first double-spindle CNC lathe, the first surface grinder, the second double-spindle CNC lathe, the first external grinder, the first centerless grinder, the second surface grinder, the third surface grinder, the third double-spindle CNC lathe, the second external grinder, and the second centerless grinder are all equipped with truss manipulators, and the column-type hydraulic press and the robotic arm polishing group are both equipped with robotic arms.

[0015] Further, the heat treatment device includes a quenching furnace and a tempering furnace.

[0016] The advantages of the present invention are as follows: 1. By constructing a closed-loop production system of "AGV intelligent logistics + robotic arm + central control system", the AGV vehicle automatically identifies the pallet code and plans the path according to the MES instruction, eliminates the collision risk caused by manual handling, and the transfer efficiency is increased by more than 60%. Each processing equipment is linked by a distributed robotic arm to realize full unmanned operation of the entire process of "processing - detection - transfer", reducing the labor cost.

[0017] 2. Based on the three-level control architecture of line detection - data linkage - dynamic compensation, the present invention can transmit the error occurring in the current process (such as the deviation of the rough turning size) to the subsequent process in real time, trigger the automatic compensation of the fine grinding allowance, and ensure the consistency of the assembly clearance through the diameter data linkage mechanism. When 3 consecutive workpieces are out of tolerance, the equipment calibration is automatically triggered to avoid batch scrapping of workpieces caused by error accumulation.

[0018] 3. In the present invention, a linear array camera is used for 360° circumferential scanning to generate a thermal map of dimensional errors, and the detection coverage rate is increased from 20% of manual random inspection to 100%. The cavity surface is scanned spirally by a multi-angle vision camera, with high defect location accuracy and improved detection efficiency. Defective parts are directed back to the designated workstations by AGV vehicles for rework, shortening the rework time. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] In order to more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the following will briefly introduce the drawings required for use in the description of the specific embodiments or the prior art. Obviously, the drawings in the following description are some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0020] Figure 1 It is a schematic flow chart of a production control method for a tablet press die of the present invention; Figure 2 It is a schematic layout diagram of the cutting process of a tablet press die production line of the present invention; Figure 3 It is a schematic layout diagram of the heat treatment and polishing processes of a tablet press die production line of the present invention; Figure 4 It is a schematic working principle diagram of an AGV vehicle; Figure 5 It is a schematic working flow chart of an AGV vehicle; Figure 6 It is a schematic working principle diagram of a robotic arm; Figure 7 It is a schematic working flow chart of a robotic arm; As Figure 2 、 Figure 3 shown, which includes: circular sawing machine 1, first double-spindle CNC lathe 2, first surface grinder 3, column-type hydraulic press 4, second double-spindle CNC lathe 5, heat treatment device 6, robotic arm polishing group 7, first cylindrical grinder 8, first centerless grinder 9, second surface grinder 10, third surface grinder 11, third double-spindle CNC lathe 12, second cylindrical grinder 13, second centerless grinder 14. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0021] The following will clearly and completely describe the technical solutions of the present invention with reference to the drawings. Obviously, the described embodiments are some, but not all, of the embodiments of the present invention. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts fall within the scope of protection of the present invention.

[0022] Embodiment 1: The present invention provides a production control method for a tablet press die, including the following steps: Step 1: Prepare the materials by sawing the workpiece raw materials into blank materials using a circular sawing machine. Place the blank materials on a pallet, and use an AGV vehicle to transport the workpiece to the first double-spindle CNC lathe for rough turning. Specifically, after the circular sawing machine cuts the materials, the operator places the blank parts on the pallet, triggers the RFID to write the process code "CT01", and the AGV vehicle goes to pick up the materials according to the MES instruction. During the movement, it dynamically avoids obstacles through a millimeter-wave radar (detection distance: 10m). The working process of the AGV vehicle includes the following steps: a. Receive instructions from the control system; b. Plan the route, identify the unique identification code of the pallet, locate the position of the pallet, and go to the specified pallet position; c. Grab the workpiece, avoid obstacles, and transport the workpiece along the preset path; d. After reaching the target position, place the workpiece on the specified tooling fixture or machine tool, and confirm that the workpiece has been correctly placed; e. Send a confirmation signal to the control system, wait to receive the next instruction from the control system, and go back to step a to continue execution.

[0023] An RFID module is configured on the above AGV vehicle. Before loading the workpiece, it conducts two-way verification with the process equipment by scanning the unique identification code of the pallet. When the identification code does not match the process instruction required for the current task of the AGV vehicle, an alarm is triggered and the exception code is uploaded to the central control system. Step 2: After the first dual-spindle lathe completes rough turning, the truss manipulator transfers the workpiece to the first surface grinder to perform the flat grinding process. A contact probe (model: Renishaw TP20) is used to perform spiral scanning. The scanning path: a workpiece with a diameter of φ50mm uses a progressive spiral line with a pitch of 3mm; data acquisition: 59 points are collected per circle, and a total of 5 circles are scanned to form a 295 sampling point cloud. The collected data is uploaded to the MES system in real time and automatically compared with the process tolerance band. When three consecutive workpieces are out of tolerance, the equipment self-test program is triggered. Specifically, the self-test program includes mechanical calibration: checking the radial runout of the lathe spindle and verifying the positioning accuracy of the probe (through standard block gauge re-test); sensor diagnosis: light source uniformity detection of the linear array camera (grayscale value standard deviation <5%), contact probe force feedback calibration (pressure value error ±0.1N). Software self-test: CNC program verification (G code syntax check), data transmission link test (ping MES server, delay <50ms). The self-check process is automatically executed: After receiving the MES command, the equipment controller calls the pre-stored self-check program (such as macro instructions). Result feedback: The self-check result (pass / fail) is uploaded to MES. If the self-check fails, MES generates a maintenance work order and assigns a technician to handle it. Resume production: After the self-check passes, MES unlocks the equipment and continues production. The system automatically re-tests 3 workpieces to confirm that the problem has been solved. Then the workpiece flatness is detected and the grinding allowance is compensated in real time. Flatness calculation: The plane equation is fitted by the least squares method to calculate the maximum deviation value Δmax. When Δmax>0.02mm, the grinder automatically adjusts the Z-axis feed amount, and the compensation amount = Δmax×1.2.

[0024] Step 3: Press the workpiece. In the pressing process, a multi-angle visual camera is used. Specifically, the multi-angle visual camera is configured as follows: 3 5-megapixel CCD cameras (FLIR BFS-U3-50S5M) are arranged in a 120° ring, with a lens focal length of 12mm and a depth of field range of ±5mm. With a blue coaxial light source, the surface of the workpiece cavity is spirally scanned. If there are cracks or deformation defects in the workpiece, the rework process is automatically triggered. Specifically, the rework process includes: Step 3-1: After the multi-angle visual camera detects a crack or deformation defect, it sends a defect code to the central control system; Step 3-2: The AGV receives the instruction and sends the defective workpiece directly from the profiling station back to the starting end of the profiling process; Step 3-3: If the defect type is crack, the pressing pressure will automatically increase by 10% and the holding time will be extended by 5 seconds; if the defect type is deformation, the local compensation pressure will be increased at the mold position corresponding to the defect area; Step 3-4: After profiling, only locally scan the defective area of the workpiece, complete the qualification determination within 3 seconds. If qualified, the workpiece enters the next process for secondary rough turning and heat treatment. If still unqualified, the workpiece is transferred to the waste area by the AGV vehicle.

[0025] Step 4: Perform secondary rough turning and heat treatment on the profiled workpiece successively. In the processes of rough turning, secondary rough turning and finish turning, use a line array camera to perform a 360° circumferential scan on the outer contour of the workpiece, and generate a dimensional error heat map through point cloud comparison.

[0026] Step 5: Perform polishing on the cavity of the heat-treated workpiece. Specifically, in the polishing process, the manipulator identifies the unique identification code of the tray, grabs the punch rod, and polishes the small head cavity. The manipulator adopts the following control methods: a. Real-time collect the normal pressure and tangential friction force on the contact surface between the grinding head and the cavity through a six-axis force sensor; b. Convert the pressure signal into a polishing depth compensation amount, and dynamically adjust the feed of the Z-axis servo motor; c. Set the gradient adjustment rule of the polishing speed according to the friction force threshold. When the friction force exceeds the safety threshold, trigger an emergency stop. Install a six-axis force sensor at the end of the manipulator. The polishing parameters are set as: the initial normal pressure Fz = 15N ± 1N; the safety friction force threshold Fx ma x = 8N. When it is detected in real time that Fz = 16.5N, the Z-axis servo motor retracts 0.05mm. If the instantaneous value of the friction force Fx reaches 7.8N, automatically reduce the spindle speed from 3000rpm to 2500rpm. The emergency braking condition is: Fx > 8N for 100ms or Fz > 20N; then successively grind the polished workpiece through the first cylindrical grinder for rough grinding and through the first centerless grinder for centerless grinding; Step 6: Grind the working length and total length of the workpiece after centerless grinding; Step 7: Finish turning the outer shape of the workpiece after surface grinding. Before finish turning the outer shape, generate a differential machining path through visual scanning. The double-spindle lathe synchronously finishes the head diameter and the punch tail. Specifically, the generation method of the visual scanning differential machining path includes: a. Obtain the three-dimensional point cloud data of the workpiece through a binocular stereo camera, and fit and align it with the CAD model; b. Generate equally spaced cross-sectional contours along the axial direction, and calculate the radial deviation Δr between the actual contour and the theoretical contour of each cross-section; c. Fit the abnormal section where Δr > 0.1mm with a B-spline curve to generate a tool compensation path; d. Decompose the tool compensation path into the main / auxiliary axis machining intervals of the double-spindle lathe, and set an overlapping cutting zone of 0.05mm to eliminate the tool mark at the joint.

[0027] Step 8: The workpiece after rough turning the outer shape is successively subjected to precision grinding of the small end by a second cylindrical grinding machine and precision grinding of the body by a second centerless grinding machine; A diameter data linkage mechanism is established between the precision grinding of the small end and the precision grinding of the body to dynamically adjust the grinding feed rate. Specifically, the diameter data linkage mechanism includes: in the precision grinding of the small end process, the diameter D1 and roundness δ1 of the small end are collected by an online roundness meter of the cylindrical grinding machine, and the data is written into an RFID tag; when the workpiece enters the precision grinding of the body process, the centerless grinding machine reads the diameter D1 and roundness δ1 of the small end, and calculates the target diameter D2 of the body section according to the formula D2 = D1 - Δ, where Δ is the process reserved assembly gap, and adjusts the fluctuation coefficient K of the grinding wheel feed speed based on the roundness δ1 value, K = 1 + 0.2× , where the upper tolerance limit (i.e., the maximum allowable tolerance value of δ1) is 0.003 mm. When precision grinding the outer diameter of the small end, the grinding reference is calibrated based on the previous surface grinding total length data, and the roundness error is verified after precision grinding; in the precision grinding of the body process, the balance state of the grinding wheel is monitored in real time by a vibration sensor, and the grinding parameters are dynamically adjusted in combination with the online roundness meter data. Specifically, the implementation scheme of the diameter compensation mechanism in the present invention is: the cylindrical grinding machine measures the diameter D1 of the small end = 25.03 mm, the roundness δ1 = 0.008 mm, calculates the target diameter D2 of the body section = 25.03 - Δ (Δ takes the process gap of 0.05 mm), monitors the balance of the grinding wheel through a vibration sensor (model: PCB 352C33), the sampling rate is 10 kHz, and when the detected vibration acceleration > 4.5 μm / s², the grinding wheel dressing program (dressing amount of diamond pen is 0.02 mm) is automatically triggered and the feed rate is adjusted to 70% of the normal value.

[0028] The present invention also provides a production line for a tablet press die production control method, which is provided with a first area, a second area, and a third area in parallel. A circular saw 1, a first double-spindle CNC lathe 2, a first surface grinder 3, a column-type hydraulic press 4, and a second double-spindle CNC lathe 5 are successively arranged in the first area; a first cylindrical grinding machine 8, a first centerless grinding machine 9, a second surface grinder 10, and a third surface grinder 11 are successively arranged in the second area; a third double-spindle CNC lathe 12, a second cylindrical grinding machine 13, and a second centerless grinding machine 14 are successively arranged in the third area; the heat treatment device 6 and the robotic arm polishing group 7 are adjacent and arranged outside the first area, the second area, and the third area. The heat treatment device 6 includes a quenching furnace and a tempering furnace. Among them, the first double-spindle CNC lathe 2, the first surface grinder 3, the second double-spindle CNC lathe 5, the first cylindrical grinding machine 8, the first centerless grinding machine 9, the second surface grinder 10, the third surface grinder 11, the third double-spindle CNC lathe 12, the second cylindrical grinding machine 13, and the second centerless grinding machine 14 are all equipped with truss manipulators, and the column-type hydraulic press 4 and the robotic arm polishing group 7 are both equipped with robotic arms.

[0029] Figure 4 It is the working principle diagram of the AGV vehicle, Figure 5 and it is the working flow chart of the AGV vehicle. As Figure 4 , Figure 5 shown, the AGV vehicle exists between various steps. It calculates the optimal driving path based on the preset map, task instructions, and real-time environmental information, and meets the task requirements of moving workpieces while avoiding obstacles. The AGV vehicle receives task instructions from the central control system or the host computer, and sequentially completes operations such as picking up goods, transporting, and unloading according to the received task instructions. The process involves interactions with external devices, realizes information exchange through I / O interfaces and communication protocols, controls the driving wheels or steering wheels of the AGV through drive motors to achieve actions such as forward, backward, and turning, and the control system precisely adjusts the motor speed and steering angle according to navigation information and path planning instructions to ensure that the AGV vehicle travels precisely along the planned path. The AGV vehicle is equipped with various safety sensors, such as lidar, ultrasonic sensors, photoelectric switches, collision bars, etc., which are used to monitor the surrounding environment in real time, detect obstacles and trigger obstacle avoidance actions. When an obstacle or potential risk is detected, the AGV vehicle will execute preset safety strategies, such as decelerating, stopping, detouring, or alarming, to ensure the safety of personnel and equipment. Through the integrated software and hardware system, the AGV trolley can realize automated and intelligent workpiece handling tasks.

[0030] In the truss, the drive system uses an AC servo motor, which drives the gear to roll with the racks fixed on the crossbeam and the ram through a worm gear reducer, thereby driving the moving parts to move rapidly along the guide rails. This drive method has the advantages of fast response speed, large acceleration, and short acceleration and deceleration times, and can meet the high-precision and high-speed motion requirements of the truss. It analyzes and processes the input signals of the sensors and the button signals by the controller. After making logical judgments, it issues execution commands to each output component, thereby realizing the precise control of the truss motion. The servo drive motor realizes the precise control of the motor speed and electromagnetic torque by receiving the instructions of the controller, thereby driving the truss to perform combined motion along the X, Y, and Z axes. In addition, the control system can also calculate the displacement and trajectory of the servo system according to the moving distance and operation rhythm of the workpiece, and dynamically adjust the PID parameters of the driver to ensure the accuracy and stability of the truss motion. The truss manipulator grabs the mold through its end effector. During the grabbing process, the control system will monitor the grabbing force in real time to ensure that the mold is stably clamped. Once the mold is accurately positioned, the truss manipulator will send it into the machining area of the lathe. After the mold enters the lathe, the control system of the lathe will start to execute the predetermined machining program to machine the mold. The control system will monitor various parameters during the machining process, such as temperature, vibration, etc., to ensure the stability and quality of the machining process. When the mold machining is completed, the control system of the lathe will send a signal to instruct the truss manipulator to take out the mold from the lathe. After the truss manipulator places the machined mold at the designated position, it will perform a reset operation to prepare for the next grabbing and machining tasks.

[0031] Figure 6 It is the working principle diagram of the robotic arm. Figure 7It is a working flowchart of a robotic arm. The robotic arm is mainly used for grasping the mold to assist the columnar hydraulic press in molding (step 4) and for polishing the mold (step 7). It consists of a main structure, a drive system, a control system, and an end effector. The main structure provides stable support and a moving path for the robotic arm; the drive system is responsible for providing power to enable the robotic arm to move along a predetermined trajectory and speed; the control system is the brain of the robotic arm, responsible for receiving commands, processing information, and sending control signals; the end effector requires a gripper when picking up the mold and a grinding head when grinding. The robotic arm is equipped with a camera and a vision sensor for real-time acquisition of the position and attitude information of the mold. This information is transmitted to the control system, which determines the precise position of the mold through image processing and algorithm analysis. Once the position of the mold is determined, the control system plans the optimal movement path of the robotic arm from the current position to the mold position according to the preset trajectory and speed. During the movement of the robotic arm towards the mold, the control system adjusts the joint angles and speed of the robotic arm in real time to ensure that the end effector can accurately reach the mold position. At the same time, the control system also monitors the contact force between the robotic arm and the mold in real time through feedback devices such as force sensors to avoid damaging the mold. When the end effector reaches the mold position, the control system controls the gripper or other clamping mechanisms to clamp the mold with appropriate force. In this process, the control system takes into account factors such as the weight, shape, and material of the mold to ensure the stability and safety of the clamping process. During the grinding process, the robotic arm needs to control the position and rotation speed of the grinding head to achieve the desired grinding effect.

[0032] In the production method of the tablet press mold production line in the present invention, the specific process is: sawing blank material → rough turning of blank material → surface grinding of blank end face → molding → rough turning → heat treatment → polishing of the cavity of the tablet shape → rough grinding of the outer diameter of the punch body by external cylindrical grinding → rough grinding of the punch body by centerless grinding → surface grinding of the working length → surface grinding of the total length → finish turning of the outer shape → fine grinding of the small end by external cylindrical grinding → fine grinding of the punch body by centerless grinding.

[0033] An automatic detection device is equipped between each process step. The size control of the grinding machine uses probe detection, and other steps all use photographic detection. After passing the on-line detection, the workpiece can be placed on the tray and the next workpiece can be picked up. After all the workpieces on the current tray are processed, they are uploaded to the control system, and the system issues commands to the AGV vehicle, and the workpiece is transported by the AGV vehicle. Specifically as follows: The raw materials are processed by the sawing machine for sawing the blanks to complete the material preparation. The materials are manually placed on the pallet and affixed with the code. The AGV car is instructed to go to the pallet through the system order. The AGV car identifies the code and moves the workpiece to the blank turning process. After the blank is processed, the AGV car is instructed to transfer the workpiece to the flat grinding end face. The manipulator identifies and grabs the blank, places the fixture, and the fixture automatically clamps and flat grinds the end face. After the flat grinding end face sequence is completed, the AGV car is instructed to move the workpiece to the profiling step. The manipulator identifies the pallet code, grabs the workpiece, places the press fixture, and the fixture locks the workpiece to perform the profiling process on the workpiece. After the profiling tooling is completed, the AGV car transfers the workpiece to the rough turning step. The truss manipulator identifies the pallet code and confirms it. Grab the workpiece and place it on the double-spindle lathe to perform rough processing on the outer diameter, head and tail of the workpiece. After the rough turning is completed, the AGV car transfers the workpiece to the heat treatment step. The heat treatment worker frames the workpiece and completes the corresponding heat treatment process. The employee places the workpiece that has completed heat treatment on a pallet and moves it to the polishing step station. The robot recognizes the pallet code, grabs the punch, and polishes the small head cavity. After polishing, the AGV moves the workpiece to the cylindrical grinding punch. The truss robot recognizes the pallet code and confirms it. Grab the workpiece, place it on the grinder, and clamp it. Rough grind the outer diameter of the workpiece. After the cylindrical rough grinding process is completed, the AGV transfers the workpiece to the centerless grinder for rough grinding. The truss robot recognizes the pallet code and confirms it. Grab the workpiece, place it on the centerless grinder, and process the outer diameter of the workpiece. After the centerless grinding process is completed, the AGV moves the workpiece to the flat grinding working length step station. The truss robot recognizes the pallet code and confirms it. Grab the workpiece, place it on the grinder, tighten the tooling, process the working length, and then flat grind the total length. Since the flat grinding working length and the flat grinding total length use the same truss, there is no need for the AGV to move the workpiece again after the processing is completed. The truss robot confirms the workpiece, places the workpiece on the grinder fixture, clamps it, and performs flat grinding for the total length. After the flat grinding for the total length is completed, the AGV vehicle moves the workpiece to the finishing step, and the truss robot identifies the pallet code and confirms it. Grab the workpiece, place it on the dual-spindle lathe, and finish the head and tail of the workpiece. After the finishing process is completed, the AGV vehicle moves the workpiece to the small head of the cylindrical grinder for finishing, and the truss robot identifies the pallet code and confirms it. Grab the workpiece, place the workpiece on the grinder, clamp it, and finish the small head of the punch rod. After the small head is finished, the AGV vehicle moves the pallet to the centerless grinding and finishing punch body sequence, and the truss robot identifies the pallet code, confirms the grabbing of the workpiece, places the workpiece on the grinder, and processes the punch rod. At this point, all the steps before the electroplating process are completed.

[0034] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the foregoing embodiments, or perform equivalent replacements on some or all of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A tablet press mold production control method, characterized in that, The following steps are involved: Step 1: Prepare the workpiece by sawing the blank with a circular saw, put it on a pallet, and use the AGV to move the workpiece to the first double-spindle CNC lathe for rough turning. Step 2: After rough turning the workpiece, perform the end surface grinding process, use a contact probe to detect its flatness and compensate the grinding allowance in real time; Step 3: Press the workpiece. During the pressing process, a multi-angle visual camera is used to spirally scan the surface of the workpiece cavity. If there are cracks or deformation defects on the workpiece, the rework process is automatically triggered. Step 4: Perform secondary rough turning and heat treatment on the workpiece after pressing; Step 5: polishing the heat-treated workpiece to obtain a sheet-shaped cavity, and then performing rough grinding on the polished workpiece by a first cylindrical grinder and centerless grinding on the first centerless grinder in sequence; Step 6: Grind the workpiece after centerless grinding to its working length and total length; Step 7: Finish the workpiece after flat grinding. Before finishing the workpiece, generate a differential machining path through visual scanning. The third dual-spindle lathe will finish the head diameter and the tail simultaneously. Step 8: The workpiece after fine turning is subjected to fine grinding of the small head by the second cylindrical grinder and fine grinding of the punch body by the second centerless grinder in sequence; Establish a diameter data linkage mechanism between the fine grinding of the small head and the fine grinding of the punch body to dynamically adjust the grinding feed rate; When fine grinding the small head externally, calibrate the grinding reference based on the total length data of the previous flat grinding, and verify the roundness error after fine grinding; In the process of fine grinding the punch body, the balance state of the grinding wheel is monitored in real time by a vibration sensor, and the grinding parameters are dynamically adjusted based on the data from the online roundness meter.

2. A tablet press mold production control method according to claim 1, characterized in that: The working process of the AGV vehicle includes the following steps: a. Receive instructions from the control system; b. Plan the route, identify the unique identification code of the pallet, locate the pallet position, and go to the designated pallet location; c. Grab the workpiece, avoid obstacles, and transfer the workpiece along the preset path; d. After reaching the target position, place the workpiece on the designated fixture or machine tool and confirm that the workpiece has been placed correctly; e. Send a confirmation signal to the control system, wait to receive the next instruction from the control system, and go to step a to continue execution.

3. A tablet press mold production control method according to claim 2, characterized in that, The AGV is equipped with a radio frequency identification module, which performs two-way verification with the process equipment by scanning the unique identification code of the pallet before loading the workpiece. When the identification code does not match the process instructions required by the task currently to be performed by the AGV, an alarm is triggered and an abnormal code is uploaded to the central control system.

4. A tablet press mold production control method according to claim 1, characterized in that: In the rough turning, secondary rough turning and finish turning processes, a linear array camera is used to perform a 360° circumferential scan of the workpiece's outer contour, and a dimensional error heat map is generated through point cloud comparison; In the process of flat grinding the end surface, the contact probe scans the end surface in a spiral path, collects flatness data of at least 50 sampling points, and calculates the grinding allowance distribution based on the least squares method to fit the plane equation; The collected data is uploaded to the MES system in real time and automatically compared with the process tolerance band. When three consecutive workpieces are out of tolerance, the equipment self-inspection program is triggered.

5. A tablet press mold production control method according to claim 1, characterized in that: In the polishing process of step 5, the robot recognizes the unique identification code of the tray, grabs the punch, and polishes the small head cavity, wherein the robot adopts the following control method: a. Use the six-dimensional force sensor to collect the normal pressure and tangential friction force of the contact surface between the grinding head and the cavity in real time; b. Convert the pressure signal into polishing depth compensation and dynamically adjust the feed rate of the Z-axis servo motor; c. Set the gradient adjustment rule of the polishing speed according to the friction threshold, and trigger an emergency stop when the friction exceeds the safety threshold.

6. A tablet press mold production control method according to claim 1, characterized in that: The method for generating the visual scanning difference processing path in step 7 includes: a. Obtain the 3D point cloud data of the workpiece through a binocular stereo camera and align it with the CAD model; b. Generate equally spaced cross-sectional profiles along the axial direction and calculate the radial deviation Δr between the actual profile of each cross-sectional profile and the theoretical profile; c. Perform B-spline curve fitting on the abnormal section with Δr>0.1mm to generate the tool compensation path; d. Decompose the tool compensation path into the main / sub-spindle machining areas of the third dual-spindle lathe, and set an overlapping cutting zone of 0.05 mm to eliminate tool marks.

7. A tablet press mold production control method according to claim 1, characterized in that: The diameter data linkage mechanism in step 8 includes: in the fine grinding process of the small head, the small head diameter D1 and the ovality δ1 are collected by the online roundness meter of the cylindrical grinder, and the data are written into the RFID tag; when the workpiece enters the fine grinding process of the punching body, the centerless grinder reads the small head diameter D1 and the ovality δ1, and calculates the target diameter D2 of the punching body section according to the formula D2=D1-Δ, where Δ is the reserved assembly gap for the process, and adjusts the fluctuation coefficient K=1+0.2× of the grinding wheel feed speed based on the ovality δ1 value. .

8. A tablet press mold production control method according to claim 1, characterized in that: The rework process in step 3 includes: Step 3-1: After the multi-angle visual camera detects cracks or deformation defects, it sends the defect code to the central control system; Step 3-2: The AGV receives the instruction and sends the defective workpiece directly from the profiling station back to the starting end of the profiling process; Step 3-3: If the defect type is crack, the pressing pressure will automatically increase by 10% and the holding time will be extended by 5 seconds; if the defect type is deformation, the local compensation pressure will be increased at the mold position corresponding to the defect area; Step 3-4: After forming, only the defective area of ​​the workpiece is scanned locally, and the qualified judgment is completed within 3 seconds. If qualified, the workpiece enters the next process for secondary rough turning and heat treatment. If it is still unqualified, the workpiece is transferred to the waste area by the AGV vehicle.

9. A production line based on the tablet press mold production control method according to claim 1, wherein a first area, a second area, and a third area are arranged in parallel, characterized in that: In the first area, a circular saw, a first double-spindle CNC lathe, a first surface grinder, a column-type hydraulic press, and a second double-spindle CNC lathe are arranged in sequence; in the second area, a first cylindrical grinder, a first centerless grinder, a second surface grinder, and a third surface grinder are arranged in sequence; in the third area, a third double-spindle CNC lathe, a second cylindrical grinder, and a second centerless grinder are arranged in sequence; the heat treatment device is adjacent to the robotic arm polishing group and is arranged outside the first area, the second area, and the third area; among them, the first double-spindle CNC lathe, the first surface grinder, the second double-spindle CNC lathe, the first cylindrical grinder, the first centerless grinder, the second surface grinder, the third surface grinder, the third double-spindle CNC lathe, the second cylindrical grinder, and the second centerless grinder are all equipped with truss robots, and the column-type hydraulic press and the robotic arm polishing group are all equipped with robotic arms.

10. The production line of a tablet press mold production control method according to claim 9, characterized in that: The heat treatment device includes a quenching furnace and a tempering furnace.

Images