Real-time control method and system for universal joint yoke semi-solid forming process

Abstract

This invention discloses a real-time control method and system for the semi-solid forming process of universal joint fork arms, relating to the field of real-time control. The method includes: setting a semi-solid forming temperature range based on the solid and liquidus temperatures of the billet, and pre-treating the billet; after polishing the mold cavity, deploying thermocouples, pressure sensors, and infrared thermal imagers in key areas to monitor the temperature field and pressure in real time; induction heating the billet to a semi-solid state, dynamically adjusting its viscosity and solid fraction, and then transporting it to the mold through a temperature-controlled pipeline; after mold closing, predicting the forming quality based on a BP neural network model and implementing multi-field coupling control; monitoring shrinkage through displacement sensors during the pressure holding stage, and initiating auxiliary heating compensation when local over-cooling occurs; and detecting dimensions and defects after demolding, and providing feedback to optimize process parameters. The advantages of this invention are: by combining real-time monitoring, dynamic control, and neural network prediction throughout the entire process with detection feedback to correct parameters, a closed loop is formed, improving the forming accuracy and stability of universal joint fork arms.

Description

Technical Field

[0001] This invention relates to the field of data processing, and in particular to a method and system for real-time control of the semi-solid forming process of universal joint fork arms. Background Technology

[0002] As a key component of the automotive transmission system, the performance of the universal joint fork arm directly affects the reliability and durability of power transmission. Traditional manufacturing processes suffer from low material utilization, insufficient forming accuracy, and uneven mechanical properties. Semi-solid forming technology, by controlling the solid-liquid ratio of the metal slurry, can achieve near-net-shape forming, reduce defects, and improve mechanical properties, making it a research hotspot in this field. However, the semi-solid forming process involves complex rheological behavior and phase transformation kinetics. Real-time fluctuations in process parameters can easily lead to porosity, segregation, or cracks inside the formed part, and traditional offline control methods are insufficient to meet the demands of high-precision production.

[0003] Current semi-solid forming methods for universal joint fork arms have several shortcomings. In terms of monitoring, most methods only deploy a few sensors at key nodes, lacking real-time capture of the temperature and pressure fields across the entire mold cavity. Furthermore, they do not incorporate infrared thermal imaging technology, making it difficult to accurately grasp the slurry filling state and prone to defects due to localized temperature and pressure imbalances. Regarding control methods, most rely on static parameters such as fixed heating power and stirring speed, lacking dynamic correction capabilities for fluctuations in slurry viscosity and solids content. Moreover, the skin effect of induction heating leads to large temperature differences between the inside and outside of the billet, resulting in poor heating uniformity. The forming process lacks intelligent prediction mechanisms, relying on experience rather than BP neural network models to predict density and dimensional accuracy, failing to proactively address multi-field coupling imbalances. During the pressure holding stage, there are often no targeted compensation methods, resulting in delayed responses to localized temperature drops exceeding 2°C / s and solidification shrinkage, leading to a high incidence of shrinkage cavities and porosity defects. In addition, most methods lack closed-loop feedback, making it difficult to automatically iterate and optimize subsequent processes based on dimensional inspection and defect detection data, resulting in low material utilization, short mold life, and poor product consistency. Summary of the Invention

[0004] To improve existing methods and systems, a real-time control method and system for the semi-solid forming process of universal joint fork arms is provided. This method uses real-time monitoring, dynamic control and neural network prediction throughout the entire process, combined with detection feedback to correct parameters to form a closed loop, ultimately significantly improving the forming accuracy and stability of universal joint fork arms, reducing defects and optimizing production efficiency.

[0005] To achieve the above objectives, the technical solution adopted by the present invention is as follows:

[0006] Real-time control methods for the semi-solid forming process of universal joint fork arms include:

[0007] Aluminum alloy blanks for universal joint fork arms are selected, and the solidus temperature and liquidus temperature of the blanks are determined by differential scanning calorimetry to determine the target temperature range for semi-solid forming. The blanks are then subjected to surface pretreatment.

[0008] Polish the cavity of the universal joint fork arm forming mold, and install K-type thermocouples and pressure sensors in key areas of the fork arm head, rod and fork root in the mold cavity. Install an infrared thermal imager on the outside of the mold cavity to capture the overall temperature field distribution of the cavity in real time.

[0009] The pretreated aluminum alloy billet is heated to the target temperature range by medium-frequency induction heating. The viscosity change and solid fraction data of the slurry are monitored in real time. When the viscosity or solid fraction deviates from the target range, the induction heating power or stirring speed is adjusted to dynamically correct the state of the slurry.

[0010] The prepared semi-solid slurry is quantitatively delivered to the mold cavity. The heating power of the heating coil wrapped around the outer wall of the delivery pipe is adjusted in real time by a PID temperature controller. An infrared temperature sensor is installed at the outlet of the delivery pipe to monitor the slurry outlet temperature in real time.

[0011] The hydraulic press is started to close the mold and apply pressure. The temperature, pressure and temperature field distribution inside the cavity are collected by thermocouples, pressure sensors and infrared thermal imagers. Based on the forming parameter prediction model of BP neural network, the forming density and dimensional accuracy data of the universal joint fork arm are obtained, and multi-field coupling control is performed based on threshold.

[0012] After the slurry fills the mold cavity, it enters the pressure holding stage. The displacement change of the hydraulic press slider is monitored in real time by the displacement sensor to indirectly determine the solidification shrinkage of the blank. If the local temperature drop rate of the cavity exceeds 2℃ / s, the auxiliary heating device of the corresponding area of ​​the mold is activated to perform pressure compensation.

[0013] After the pressure holding period, demolding is performed. A coordinate measuring machine is used to detect the key dimensions of the universal joint fork arm in real time. An ultrasonic flaw detector is used to detect defects inside the workpiece and correct the temperature range, pressure, or pressure holding time parameters for the next forming process.

[0014] Preferably, the step of selecting aluminum alloy blanks for universal joint fork arms, determining the solidus and liquidus temperatures of the blanks using differential scanning calorimetry to identify the target temperature range for semi-solid forming, and performing surface pretreatment on the blanks specifically includes:

[0015] Aluminum alloy blanks for universal joint fork arms are selected, and differential scanning calorimetry is used to test the selected blanks to determine the solidus temperature and liquidus temperature of the blanks. The target temperature range for semi-solid forming is determined based on the test results.

[0016] The oxide scale and oil stains on the surface of the blank are removed by physical grinding. Sandpaper is used to grind the surface of the blank until the surface roughness Ra≤1.6μm.

[0017] The surface-treated billet is placed in a medium-frequency induction heating furnace and preheated using a segmented heating mode.

[0018] Preferably, the polishing treatment of the universal joint fork arm forming mold cavity, the placement of K-type thermocouples and pressure sensors in key areas of the fork arm head, rod, and fork root of the mold cavity, and the installation of an infrared thermal imager on the outside of the mold cavity to capture the overall temperature field distribution of the cavity in real time specifically include:

[0019] The universal joint fork arm forming mold is fixed on the hydraulic press worktable, and the inner wall of the mold cavity is polished with diamond polishing paste. The surface roughness is repeatedly polished until Ra≤0.8μm.

[0020] K-type thermocouples and pressure sensors are installed at the head of the cavity fork arm, the root of the fork, and the rod. The sensors are fixed by high-temperature resistant ceramic insulating sleeves.

[0021] A special support is built on the outside of the mold cavity, and the infrared thermal imager is fixed on the support. The lens angle is adjusted to clearly capture the overall temperature field distribution of the cavity.

[0022] Preferably, the step of heating the pretreated aluminum alloy billet to the target temperature range via medium-frequency induction heating, and monitoring the viscosity change and solid fraction data of the slurry in real time, and adjusting the induction heating power or stirring speed to dynamically correct the slurry state when the viscosity or solid fraction deviates from the target range, specifically includes:

[0023] The pretreated aluminum alloy billet is fed into the stirring chamber. The medium frequency induction heating parameters are set, the heating is started, the stirring power and stirring speed are set, and the heating and stirring are started simultaneously.

[0024] The billet temperature is gradually increased to the target temperature range by medium-frequency induction heating, and the heating status is monitored in real time during the process.

[0025] Insert the rotational viscometer probe into the slurry to monitor the viscosity. If the viscosity exceeds 2000 Pa·s, increase the induction heating power. If it is below 500 Pa·s, decrease the heating power or increase the stirring speed.

[0026] At a position 50-100 mm horizontally from the viscometer probe, measure the solid fraction using a laser particle size analyzer. If it deviates from the target range, adjust the heating power or stirring speed.

[0027] Preferably, the step of quantitatively conveying the prepared semi-solid slurry to the mold cavity, adjusting the heating power of the heating coil wound around the outer wall of the conveying pipe in real time using a PID temperature controller, and installing an infrared temperature sensor at the outlet of the conveying pipe to monitor the slurry outlet temperature in real time specifically includes:

[0028] Based on the volume of the universal joint fork arm, the required slurry volume is calculated, and the matching parameters of the conveying flow rate, conveying speed and screw speed of the servo screw conveyor are set.

[0029] An infrared temperature sensor is installed at the outlet of the conveying pipeline, and the angle of the sensor is adjusted to face the slurry outlet to capture the slurry temperature in real time.

[0030] During the semi-solid slurry transportation process, the pipeline temperature is monitored in real time by a PID controller. Combined with the outlet temperature feedback from the infrared temperature probe, if the slurry temperature loss exceeds 5°C or the outlet temperature is lower than the lower limit of the target range, the power of the heating coil is increased. If the outlet temperature is higher than the upper limit of the target range, the power of the heating coil is reduced.

[0031] Preferably, the process of starting the hydraulic press to close the mold and apply pressure involves collecting temperature, pressure, and temperature field distribution within the mold cavity using thermocouples, pressure sensors, and an infrared thermal imager. Based on a BP neural network-based forming parameter prediction model, the predicted forming density and dimensional accuracy data of the universal joint fork arm are obtained. Furthermore, multi-field coupling control based on threshold values ​​specifically includes:

[0032] Start the hydraulic press to drive the upper mold to move downwards and close the mold. When the mold is closed to the preset position, switch from the mold closing mode to the pressurization mode.

[0033] The temperature and pressure data of the fork head, rod, and fork root are collected in real time by K-type thermocouples and pressure sensors installed in the cavity, and the overall temperature field distribution data of the cavity is captured by an infrared thermal imager.

[0034] By inputting the collected data into the pre-trained BP neural network model, the forming density and dimensional accuracy data of the universal joint fork arm are obtained. Based on the difference between the predicted data and the preset threshold, the forming process is controlled by adjusting the power of the auxiliary heating device or controlling the pressure of the hydraulic press.

[0035] Preferably, after the slurry fills the mold cavity, the pressure holding stage begins. A displacement sensor monitors the displacement change of the hydraulic press slider in real time to indirectly determine the solidification shrinkage of the blank. If the rate of temperature drop in the local area of ​​the cavity exceeds 2°C / s, the auxiliary heating device for the corresponding area of ​​the mold is activated for pressure compensation. Specifically, this includes:

[0036] When the mold cavity pressure reaches the preset threshold, the system switches from pressurization mode to pressure holding mode, monitors displacement sensor data, and if the slider displacement changes by more than 0.1mm, increases the pressure holding pressure and extends the pressure holding time to compensate for shrinkage.

[0037] The temperature drop rate of each area is monitored in real time by a type K thermocouple. If the temperature drop rate of a certain area exceeds 2℃ / s, the ceramic heating element in that area is activated to slow down the temperature drop.

[0038] After adjustment, continuously monitor pressure, displacement, and temperature data until the pressure holding time ends.

[0039] Preferably, after the pressure holding period, demolding is performed, and a coordinate measuring machine is used to detect the key dimensions of the universal joint fork arm in real time. An ultrasonic flaw detector is used to detect defects inside the workpiece, and the temperature range, pressure, or pressure holding time parameters for the next forming are adjusted accordingly. Specifically, this includes:

[0040] After the pressure holding stage is completed, switch to demolding mode and observe the fit between the workpiece and the cavity. If slight adhesion occurs, fine-tune the demolding speed.

[0041] After demolding, the universal joint fork arm is fixed on the worktable of the coordinate measuring machine. Key dimensions such as fork spacing, rod diameter, and head thickness are selected for inspection, and the deviations between the actual values ​​and the design values ​​of each dimension are recorded.

[0042] The head and root area of ​​the workpiece are scanned by an ultrasonic flaw detector to detect whether there are shrinkage cavities or shrinkage porosity inside, and the location and size of the defects are marked.

[0043] Based on dimensional deviations exceeding thresholds or the presence of defects exceeding standards, the temperature range, pressure, and holding time for the next forming process are adjusted to form a closed-loop control.

[0044] Furthermore, a real-time control system for the semi-solid forming process of the universal joint fork arm is proposed, including:

[0045] Mold condition monitoring module: By polishing the mold cavity and installing thermocouples, pressure sensors and infrared thermal imagers, it can collect real-time data on cavity temperature, pressure and temperature field distribution.

[0046] Slurry state control module: Controls slurry viscosity and solid fraction through medium-frequency induction heating and stirring, and dynamically adjusts heating power or stirring speed to maintain target parameters;

[0047] Forming process prediction and control module: Based on the BP neural network model, it predicts the forming density and dimensional accuracy in real time, and optimizes the process parameters through multi-field coupling control;

[0048] Pressure holding compensation module: Monitors slider displacement and local temperature changes, dynamically adjusts pressure holding pressure or activates auxiliary heating device to compensate for contraction and sudden temperature drop;

[0049] Quality inspection feedback module: By detecting the size and internal defects of the workpiece, the feedback data is used to correct the forming parameters for the next time, realizing automatic iterative optimization of process parameters and closed-loop control of the whole process;

[0050] Processor: The processor is used to handle the calculation process of each formula and the construction calculation process of each model.

[0051] Compared with the prior art, the advantages of the present invention are:

[0052] By deploying sensors and infrared thermal imagers in key areas of the mold, the cavity temperature, pressure, and overall temperature field are captured in real time. Combined with real-time monitoring of viscosity and solids content during the slurry heating stage, the induction heating power, stirring speed, and heat transfer power can be dynamically adjusted to prevent the slurry state and temperature from deviating from the target range. During mold closing and pressurization, a BP neural network model can predict the forming density and dimensional accuracy in advance and carry out multi-field coupling control. During the pressure holding stage, displacement monitoring can compensate for solidification shrinkage, and auxiliary heating can be activated to suppress excessive local temperature drop, effectively reducing defects such as shrinkage cavities and porosity. After demolding, dimensional inspection and defect detection feedback data are used to correct subsequent process parameters, forming a closed-loop control throughout the entire process. Ultimately, this significantly improves the dimensional accuracy, forming stability, and product consistency of the universal joint fork arm, while also achieving automatic iterative optimization of process parameters, reducing production losses and improving production efficiency. Attached Figure Description

[0053] Figure 1 This is a schematic diagram of the method proposed in this invention;

[0054] Figure 2 This is a schematic diagram of obtaining the target temperature range and surface pretreatment proposed in this invention;

[0055] Figure 3 This is a schematic diagram of real-time data acquisition proposed in this invention;

[0056] Figure 4 This is a schematic diagram of the dynamic correction of slurry state proposed in this invention;

[0057] Figure 5 This is a schematic diagram of the quantitative slurry delivery method proposed in this invention;

[0058] Figure 6 This is a schematic diagram of the multi-field coupling control proposed in this invention;

[0059] Figure 7 This is a schematic diagram of the pressure compensation proposed in this invention;

[0060] Figure 8This is a schematic diagram of the defect detection method proposed in this invention. Detailed Implementation

[0061] The following description is intended to disclose the invention and enable those skilled in the art to implement it. The preferred embodiments described below are merely examples, and other obvious variations will occur to those skilled in the art.

[0062] The real-time control system for the semi-solid forming process of the universal joint fork arm includes:

[0063] Mold condition monitoring module: By polishing the mold cavity and installing thermocouples, pressure sensors and infrared thermal imagers, it can collect real-time data on cavity temperature, pressure and temperature field distribution.

[0064] Slurry state control module: Controls slurry viscosity and solid fraction through medium-frequency induction heating and stirring, and dynamically adjusts heating power or stirring speed to maintain target parameters;

[0065] Forming process prediction and control module: Based on the BP neural network model, it predicts the forming density and dimensional accuracy in real time, and optimizes the process parameters through multi-field coupling control;

[0066] Pressure holding compensation module: Monitors slider displacement and local temperature changes, dynamically adjusts pressure holding pressure or activates auxiliary heating device to compensate for contraction and sudden temperature drop;

[0067] Quality inspection feedback module: By detecting the size and internal defects of the workpiece, the feedback data is used to correct the forming parameters for the next time, realizing automatic iterative optimization of process parameters and closed-loop control of the whole process;

[0068] Processor: The processor is used to handle the calculation process of each formula and the construction calculation process of each model.

[0069] See Figure 1 As shown, the real-time control method for the semi-solid forming process of the universal joint fork arm includes:

[0070] Step 1: Select aluminum alloy blanks for universal joint fork arms, determine the solidus temperature and liquidus temperature of the blanks using differential scanning calorimetry, determine the target temperature range for semi-solid forming, and perform surface pretreatment on the blanks.

[0071] Step 2: Polish the cavity of the universal joint fork arm forming mold, and install K-type thermocouples and pressure sensors in key areas such as the fork arm head, rod and fork root of the mold cavity. Install an infrared thermal imager on the outside of the mold cavity to capture the overall temperature field distribution of the cavity in real time.

[0072] Step 3: The pretreated aluminum alloy billet is heated to the target temperature range by medium-frequency induction heating. The viscosity change and solid fraction data of the slurry are monitored in real time. When the viscosity or solid fraction deviates from the target range, the induction heating power or stirring speed is adjusted to dynamically correct the state of the slurry.

[0073] Step 4: The prepared semi-solid slurry is quantitatively delivered to the mold cavity. The heating power of the heating coil wrapped around the outer wall of the delivery pipe is adjusted in real time by the PID temperature controller. An infrared temperature sensor is installed at the outlet of the delivery pipe to monitor the slurry outlet temperature in real time.

[0074] Step 5: Start the hydraulic press to close the mold and apply pressure. Collect the temperature, pressure and temperature field distribution in the cavity through thermocouples, pressure sensors and infrared thermal imagers. Based on the forming parameter prediction model of BP neural network, obtain the predicted forming density and dimensional accuracy data of the universal joint fork arm, and perform multi-field coupling control based on threshold.

[0075] Step 6: After the slurry fills the mold cavity, it enters the pressure holding stage. The displacement change of the hydraulic press slider is monitored in real time by the displacement sensor to indirectly determine the solidification shrinkage of the billet. If the local temperature drop rate of the cavity exceeds 2℃ / s, the auxiliary heating device of the corresponding area of ​​the mold is activated to perform pressure compensation.

[0076] Step 7: After the pressure holding is completed, demolding is performed. A coordinate measuring machine is used to detect the key dimensions of the universal joint fork arm in real time. An ultrasonic flaw detector is used to detect defects inside the workpiece and correct the temperature range, pressure, or pressure holding time parameters for the next forming.

[0077] See Figure 2 As shown, aluminum alloy blanks for universal joint fork arms are selected. The solidus temperature and liquidus temperature of the blanks are measured using a differential scanning calorimeter to determine the target temperature range for semi-solid forming. Surface pretreatment of the blanks specifically includes:

[0078] Aluminum alloy blanks for universal joint fork arms are selected, and differential scanning calorimetry is used to test the selected blanks to determine the solidus temperature and liquidus temperature of the blanks. The target temperature range for semi-solid forming is determined based on the test results.

[0079] The oxide scale and oil stains on the surface of the blank are removed by physical grinding. Sandpaper is used to grind the surface of the blank until the surface roughness Ra≤1.6μm.

[0080] The surface-treated billet is placed in a medium-frequency induction heating furnace and preheated using a segmented heating mode.

[0081] Specifically, observe the temperature-heat flow curve changes in real time. When the curve shows the first obvious inflection point, record the corresponding temperature as the solidus line Ts. Continue heating until the curve shows the second inflection point, and record the corresponding temperature as the liquidus line Tl. If abnormal fluctuations occur in the curve during the process, the experiment must be stopped, and the sample must be checked for detachment or argon flow rate for stability. After troubleshooting, the test should be repeated. Based on Ts and Tl, the target temperature range for semi-solid forming is determined as [Ts+ΔT1, Tl-ΔT2], where ΔT1=5-15℃ and ΔT2=8-20℃. Adjust the parameters for different alloys.

[0082] The pre-treated billet is gently placed on the support inside the furnace. The first stage of heating is as follows: the power of the medium frequency induction heating furnace is set to 3kW, the heating rate is 5-10℃ / min, and the billet temperature is raised to 150℃. After reaching the target temperature, it is held for 10 minutes. The purpose of this stage is to release the internal stress generated in the billet during the rolling process.

[0083] Second stage of heating: Adjust the heating rate to 3-5℃ / min and continue heating to 200-300℃; During the heating process, use a thermocouple inserted into the center of the billet to monitor the core temperature in real time to ensure that the temperature difference between the surface and the core is ≤5℃. If the temperature difference exceeds the limit, reduce the heating rate by 0.5℃ / min.

[0084] Heat preservation stage: After reaching the target preheating temperature, maintain the temperature for 15-30 minutes. During the heat preservation period, the heating power will be automatically adjusted to 1-2kW to maintain a stable temperature. Record the surface and core temperatures every 5 minutes, and the deviation must always be ≤±3℃.

[0085] See Figure 3 As shown, the cavity of the universal joint fork arm forming mold is polished. K-type thermocouples and pressure sensors are installed in key areas of the fork arm head, rod, and fork root in the mold cavity. An infrared thermal imager is installed on the outside of the mold cavity to capture the overall temperature field distribution of the cavity in real time. Specifically, this includes:

[0086] The universal joint fork arm forming mold is fixed on the hydraulic press worktable, and the inner wall of the mold cavity is polished with diamond polishing paste. The surface roughness is repeatedly polished until Ra≤0.8μm.

[0087] K-type thermocouples and pressure sensors are installed at the head of the cavity fork arm, the root of the fork, and the rod. The sensors are fixed by high-temperature resistant ceramic insulating sleeves.

[0088] A special support is built on the outside of the mold cavity, and the infrared thermal imager is fixed on the support. The lens angle is adjusted to clearly capture the overall temperature field distribution of the cavity.

[0089] See Figure 4As shown, the pretreated aluminum alloy billet is heated to the target temperature range by medium-frequency induction heating. The viscosity change and solids content of the slurry are monitored in real time. When the viscosity or solids content deviates from the target range, the induction heating power or stirring speed is adjusted to dynamically correct the slurry state. Specifically, this includes:

[0090] The pretreated aluminum alloy billet is fed into the stirring chamber. The medium frequency induction heating parameters are set, the heating is started, the stirring power and stirring speed are set, and the heating and stirring are started simultaneously.

[0091] The billet temperature is gradually increased to the target temperature range by medium-frequency induction heating, and the heating status is monitored in real time during the process.

[0092] Insert the rotational viscometer probe into the slurry to monitor the viscosity. If the viscosity exceeds 2000 Pa·s, increase the induction heating power. If it is below 500 Pa·s, decrease the heating power or increase the stirring speed.

[0093] At a position 50-100 mm horizontally from the viscometer probe, measure the solid fraction using a laser particle size analyzer. If it deviates from the target range, adjust the heating power or stirring speed.

[0094] Specifically, the first stage of heating is set with a medium-frequency induction heating power of 3-5kW and a heating rate of 5℃ / min, raising the temperature from room temperature to Ts-50℃; during this stage, the stirring device is started, with a speed of 100-150r / min and a stirring direction of clockwise to ensure that the billet is heated evenly.

[0095] The heating rate is reduced to 2℃ / min in the second stage, and heating continues to the target temperature range [Ts+ΔT1, Tl-ΔT2]. During this stage, the stirring speed is increased to 200-300r / min to ensure that the solid phase grains in the slurry are fully broken and spheroidized.

[0096] During the process, it is necessary to avoid heating too quickly, which would result in too much liquid phase, or stirring too slowly, which would result in coarse grains. The key control points are: temperature and stirring are linked. When heating reaches Ts+ΔT1, stop heating, keep stirring for 10 minutes, and then continue heating to the midpoint of the interval. If the temperature sensor of the device shows a sudden temperature rise during the heating process, immediately reduce the heating power by 0.5-1kW and increase the stirring speed by 20-30r / min to balance the temperature through heat dissipation by stirring.

[0097] If the viscosity exceeds 2000 Pa·s: immediately increase the medium-frequency induction heating power by 0.5-1 kW, keep the stirring speed constant, until it drops to 1800-2000 Pa·s; if the viscosity does not decrease after heating, check if the billet has softened completely.

[0098] For viscosity below 500 Pa·s: reduce heating power by 0.5-1 kW, maintain the rotation speed or increase stirring speed by 20-50 r / min, increase the viscosity by stirring to increase the contact probability of solid phase grains, until it reaches 500-700 Pa·s;

[0099] When the slurry viscosity deviates from the target range of 500~2000 Pa·s, the slurry fluidity is maintained by adjusting the medium-frequency induction heating power. The dynamic correction formula for slurry viscosity-heating power is:

[0100]

[0101] in, This is the corrected medium-frequency induction heating power. This is the initial heating power. This is the viscosity correction factor. The target viscosity of the slurry. This refers to the slurry viscosity, which is monitored in real time.

[0102] See Figure 5 As shown, the prepared semi-solid slurry is quantitatively conveyed to the mold cavity. The heating power of the heating coil wound around the outer wall of the conveying pipe is adjusted in real time by a PID temperature controller. An infrared temperature sensor is installed at the outlet of the conveying pipe to monitor the slurry outlet temperature in real time. Specifically, this includes:

[0103] Based on the volume of the universal joint fork arm, the required slurry volume is calculated, and the matching parameters of the conveying flow rate, conveying speed and screw speed of the servo screw conveyor are set.

[0104] An infrared temperature sensor is installed at the outlet of the conveying pipeline, and the angle of the sensor is adjusted to face the slurry outlet to capture the slurry temperature in real time.

[0105] During the semi-solid slurry transportation process, the pipeline temperature is monitored in real time by a PID controller. Combined with the outlet temperature feedback from the infrared temperature probe, if the slurry temperature loss exceeds 5°C or the outlet temperature is lower than the lower limit of the target range, the power of the heating coil is increased. If the outlet temperature is higher than the upper limit of the target range, the power of the heating coil is reduced.

[0106] Specifically, three K-type thermocouples are evenly arranged along the length of the conveying pipeline, located at the feed end, middle, and discharge end, respectively, to collect the temperature of the inner wall of the pipeline in real time; the temperature data is transmitted to the PID controller every 1 second to form a temperature distribution curve, which is convenient for judging the heating uniformity.

[0107] When the feed temperature is below 590℃, the PID controller automatically increases the power of the heating coil. After heating for 5 seconds, the temperature change is observed. If the temperature rises above 595℃, the current power is maintained. If the temperature in the middle is above 620℃, the power is reduced to 25% to avoid local overheating and excessive liquid phase in the slurry. Through continuous dynamic adjustment, the temperature fluctuation at each point in the pipeline is ensured to be ≤±3℃, providing a basis for stable slurry temperature.

[0108] The infrared temperature probe collects the slurry outlet temperature every 1 second, and the data is synchronously transmitted to the central control unit and PID controller. When the outlet temperature is 586℃, the system determines that temperature compensation needs to be strengthened. The PID controller immediately increases the power of the heating coil to 50% and extends the coil heating time. If the outlet temperature rises to 590℃, the power is restored to 30% to maintain temperature stability.

[0109] If the outlet temperature drops suddenly, rule out a fault in the heating coil; if the slurry feed temperature is too low, the central control unit sends a signal to the preparation device to prompt an increase in the slurry preparation temperature by 5°C. At the same time, the heating power at the conveying end is increased to 60%, providing double compensation for temperature loss and ensuring that the outlet temperature quickly returns to the target range.

[0110] See Figure 6 As shown, the hydraulic press is started to close the mold and apply pressure. Temperature, pressure, and temperature field distribution within the cavity are collected using thermocouples, pressure sensors, and an infrared thermal imager. Based on a BP neural network-based forming parameter prediction model, the predicted forming density and dimensional accuracy data of the universal joint fork arm are obtained. Multi-field coupling control is then performed based on thresholds, specifically including:

[0111] Start the hydraulic press to drive the upper mold to move downwards and close the mold. When the mold is closed to the preset position, switch from the mold closing mode to the pressurization mode.

[0112] The temperature and pressure data of the fork head, rod, and fork root are collected in real time by K-type thermocouples and pressure sensors installed in the cavity, and the overall temperature field distribution data of the cavity is captured by an infrared thermal imager.

[0113] By inputting the collected data into the pre-trained BP neural network model, the forming density and dimensional accuracy data of the universal joint fork arm are obtained. Based on the difference between the predicted data and the preset threshold, the forming process is controlled by adjusting the power of the auxiliary heating device or controlling the pressure of the hydraulic press.

[0114] Specifically, the pressurization system is activated, increasing the pressure from 0 to 50 MPa at a rate of 5 MPa / s. During this process, the cavity collects temperature, pressure, and overall temperature field data, forming a temperature-pressure-time linkage curve. If the infrared thermal imager shows that the temperature at the fork root is lower than other areas, it is determined that the temperature at the fork root is insufficient. The auxiliary heating device in that area is immediately activated, and the pressurization pressure in the corresponding area at the fork root is slightly increased using a hydraulic press. After heating for 3 seconds, the thermocouple feedback indicates that the temperature has risen to 586℃ and the pressure has stabilized at 55 MPa, indicating that the control is effective and the current parameters are maintained. If the temperature still does not rise, the heating element wiring is checked, reconnected, and then the control is adjusted again.

[0115] When the pressure reaches 80 MPa, the system switches to the high-pressure stage. Real-time temperature and pressure data are input into the BP neural network model, which outputs predicted forming density and dimensional accuracy deviation values, compared to the preset acceptable range. If the model predicts a forming density of 2.65 g / cm³ < 2.68 g / cm³, and the pressure sensor reports a rod pressure of 72 MPa, it is determined that insufficient rod pressure leads to poor compaction. Only the hydraulic press pressure is increased, without initiating heating. After the pressure reaches 95 MPa, the model predicts the density gradually increases to 2.69 g / cm³, stabilizing after 3 seconds. Pressurization is then stopped, and the pressure is maintained at 95 MPa. If the predicted density still does not meet the standard, the pressure is further increased to 100 MPa, while the pressurization time is extended by 2 seconds.

[0116] The infrared thermal imager continuously captures the overall temperature of the cavity. If the head temperature is found to be 620℃, although it has not reached the control threshold of below 5℃, it may cause excessive liquid phase in the head. The cooling mode of the head auxiliary heating device is activated, and the head temperature drops to 612℃ after 3 seconds, returning to the safe range. During this process, the pressure is monitored simultaneously to ensure that the temperature adjustment does not cause pressure fluctuations.

[0117] See Figure 7 As shown, after the slurry fills the mold cavity, it enters the pressure holding stage. The displacement change of the hydraulic press slider is monitored in real time by a displacement sensor to indirectly determine the solidification shrinkage of the blank. If the local temperature drop rate in the cavity exceeds 2℃ / s, the auxiliary heating device in the corresponding area of ​​the mold is activated for pressure compensation. Specific conditions include:

[0118] When the mold cavity pressure reaches the preset threshold, the system switches from pressurization mode to pressure holding mode, monitors displacement sensor data, and if the slider displacement changes by more than 0.1mm, increases the pressure holding pressure and extends the pressure holding time to compensate for shrinkage.

[0119] The temperature drop rate of each area is monitored in real time by a type K thermocouple. If the temperature drop rate of a certain area exceeds 2℃ / s, the ceramic heating element in that area is activated to slow down the temperature drop.

[0120] After adjustment, continuously monitor pressure, displacement, and temperature data until the pressure holding time ends.

[0121] Specifically, the cavity thermocouple collects temperature data every 1 second, automatically calculates the temperature difference between two adjacent data collections, divides it by the time interval to obtain the temperature drop rate, and then initiates auxiliary heating in stages according to the cooling rate.

[0122] Slight cooling: This is considered normal heat dissipation. Start low-power heating and re-measure the cooling rate after 5 seconds. If the cooling rate drops below 1℃ / s, maintain the current power. If it still exceeds 1.5℃ / s, increase the power by 0.5kW to avoid excessive heating and excessive liquid phase.

[0123] Moderate cooling: If the cooling rate is determined to be too fast, start medium power heating, check the mold cooling water circuit, and avoid continuous heat dissipation; retest after heating for 3 seconds. If the cooling rate drops to below 1.5℃ / s, reduce the power to 0.5-1kW and maintain it.

[0124] Severe cooling: If determined to be abnormal heat dissipation, start high-power heating, check if there is a gap at the contact point between the mold and the worktable. If there is a gap, fill it with a high-temperature resistant gasket. After the cooling rate drops to below 2℃ / s, gradually reduce the heating power.

[0125] See Figure 8 As shown, after the pressure holding period, demolding is performed. A coordinate measuring machine is used to detect the key dimensions of the universal joint fork arm in real time, and an ultrasonic flaw detector is used to detect defects inside the workpiece. The temperature range, pressure, or holding time parameters for the next forming process are then adjusted. Specifically, this includes:

[0126] After the pressure holding stage is completed, switch to demolding mode and observe the fit between the workpiece and the cavity. If slight adhesion occurs, fine-tune the demolding speed.

[0127] After demolding, the universal joint fork arm is fixed on the worktable of the coordinate measuring machine. Key dimensions such as fork spacing, rod diameter, and head thickness are selected for inspection, and the deviations between the actual values ​​and the design values ​​of each dimension are recorded.

[0128] The head and root area of ​​the workpiece are scanned by an ultrasonic flaw detector to detect whether there are shrinkage cavities or shrinkage porosity inside, and the location and size of the defects are marked.

[0129] Based on dimensional deviations exceeding thresholds or the presence of defects exceeding standards, the temperature range, pressure, and holding time for the next forming process are adjusted to form a closed-loop control.

[0130] Specifically, dimensional inspection and defect inspection data are categorized into: qualified data, dimensional deviation data, and defect exceeding standards data. A deviation-cause correlation algorithm is used, combined with previous forming parameters, to analyze the causes of deviations.

[0131] If the fork spacing is too small and the temperature at the root of the fork is too low during the pressure holding stage, it is determined that the temperature during the pressure holding stage is insufficient, resulting in excessive shrinkage; if there is a 0.6mm defect at the head and the solid fraction of the semi-solid slurry is too low, it is determined that the solid fraction of the slurry is insufficient, resulting in shrinkage during solidification; if the diameter of the rod is too large and the pressure of the rod is too high during the pressure application stage, it is determined that the pressure application is too high, resulting in overfilling.

[0132] Based on the cause of the deviation, the key parameters for the next forming are corrected, and the correction range follows the principle of small-step fine-tuning and precise matching: if the dimensional deviation is caused by insufficient temperature, the upper limit of the semi-solid target temperature range is increased by 2-3℃, and the auxiliary heating time of the holding pressure stage is extended by 2-3s; if the defect is caused by excessive temperature, the lower limit of the range is reduced by 2℃, and the liquid phase ratio is reduced.

[0133] If the dimensions are out of tolerance due to insufficient pressure, increase the pressure in the corresponding area by 5-8 MPa; if the dimensions are too large due to excessive pressure, decrease the pressure in the corresponding area by 5-7 MPa.

[0134] If excessive shrinkage occurs due to insufficient holding time, extend the holding time by 3-5 seconds; if excessive holding time causes the workpiece to become too brittle, shorten the holding time by 2-3 seconds.

[0135] It should be noted that the order of the above embodiments of the present invention is merely for descriptive purposes and does not represent the superiority or inferiority of the embodiments. Furthermore, the above description focuses on specific embodiments of this specification. Additionally, the processes depicted in the accompanying drawings do not necessarily require a specific or sequential order to achieve the desired results. In some embodiments, multitasking and parallel processing are possible or may be advantageous.

[0136] The various embodiments in this specification are described in a progressive manner. The same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on describing the differences from other embodiments.

[0137] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for real-time control of the semi-solid forming process of a universal joint fork arm, characterized in that, include: Aluminum alloy blanks for universal joint fork arms are selected, and the solidus temperature and liquidus temperature of the blanks are determined by differential scanning calorimetry to determine the target temperature range for semi-solid forming. The blanks are then subjected to surface pretreatment. Polish the cavity of the universal joint fork arm forming mold, and install K-type thermocouples and pressure sensors in key areas of the fork arm head, rod and fork root in the mold cavity. Install an infrared thermal imager on the outside of the mold cavity to capture the overall temperature field distribution of the cavity in real time. The pretreated aluminum alloy billet is heated to the target temperature range by medium-frequency induction heating. The viscosity change and solid fraction data of the slurry are monitored in real time. When the viscosity or solid fraction deviates from the target range, the induction heating power or stirring speed is adjusted to dynamically correct the state of the slurry. The prepared semi-solid slurry is quantitatively delivered to the mold cavity. The heating power of the heating coil wrapped around the outer wall of the delivery pipe is adjusted in real time by a PID temperature controller. An infrared temperature sensor is installed at the outlet of the delivery pipe to monitor the slurry outlet temperature in real time. The hydraulic press is started to close the mold and apply pressure. The temperature, pressure and temperature field distribution inside the cavity are collected by thermocouples, pressure sensors and infrared thermal imagers. Based on the forming parameter prediction model of BP neural network, the forming density and dimensional accuracy data of the universal joint fork arm are obtained, and multi-field coupling control is performed based on threshold. After the slurry fills the mold cavity, it enters the pressure holding stage. The displacement change of the hydraulic press slider is monitored in real time by the displacement sensor to indirectly determine the solidification shrinkage of the blank. If the local temperature drop rate of the cavity exceeds 2℃ / s, the auxiliary heating device of the corresponding area of ​​the mold is activated to perform pressure compensation. After the pressure holding period, demolding is performed. A coordinate measuring machine is used to detect the key dimensions of the universal joint fork arm in real time. An ultrasonic flaw detector is used to detect defects inside the workpiece and correct the temperature range, pressure, or pressure holding time parameters for the next forming process.

2. The real-time control method for the semi-solid forming process of the universal joint fork arm according to claim 1, characterized in that, The process of selecting aluminum alloy blanks for universal joint fork arms, determining the solidus and liquidus temperatures of the blanks using differential scanning calorimetry to identify the target temperature range for semi-solid forming, and performing surface pretreatment on the blanks specifically includes: Aluminum alloy blanks for universal joint fork arms are selected, and differential scanning calorimetry is used to test the selected blanks to determine the solidus temperature and liquidus temperature of the blanks. The target temperature range for semi-solid forming is determined based on the test results. The oxide scale and oil stains on the surface of the blank are removed by physical grinding. Sandpaper is used to grind the surface of the blank until the surface roughness Ra≤1.6μm. The surface-treated billet is placed in a medium-frequency induction heating furnace and preheated using a segmented heating mode.

3. The real-time control method for the semi-solid forming process of the universal joint fork arm according to claim 1, characterized in that, The process of polishing the mold cavity of the universal joint fork arm forming mold, and installing K-type thermocouples and pressure sensors in key areas such as the fork arm head, rod, and fork root of the mold cavity, and installing an infrared thermal imager on the outside of the mold cavity to capture the overall temperature field distribution of the cavity in real time, specifically includes: The universal joint fork arm forming mold is fixed on the hydraulic press worktable, and the inner wall of the mold cavity is polished with diamond polishing paste. The surface roughness is repeatedly polished until Ra≤0.8μm. K-type thermocouples and pressure sensors are installed at the head of the cavity fork arm, the root of the fork, and the rod. The sensors are fixed by high-temperature resistant ceramic insulating sleeves. A special support is built on the outside of the mold cavity, and the infrared thermal imager is fixed on the support. The lens angle is adjusted to clearly capture the overall temperature field distribution of the cavity.

4. The real-time control method for the semi-solid forming process of the universal joint fork arm according to claim 1, characterized in that, The process of heating the pretreated aluminum alloy billet to a target temperature range via medium-frequency induction heating, and monitoring the viscosity change and solid fraction data of the slurry in real time, dynamically correcting the slurry state by adjusting the induction heating power or stirring speed when the viscosity or solid fraction deviates from the target range, specifically includes: The pretreated aluminum alloy billet is fed into the stirring chamber. The medium frequency induction heating parameters are set, the heating is started, the stirring power and stirring speed are set, and the heating and stirring are started simultaneously. The billet temperature is gradually increased to the target temperature range by medium-frequency induction heating, and the heating status is monitored in real time during the process. Insert the rotational viscometer probe into the slurry to monitor the viscosity. If the viscosity exceeds 2000 Pa·s, increase the induction heating power. If it is below 500 Pa·s, decrease the heating power or increase the stirring speed. At a position 50-100 mm horizontally from the viscometer probe, measure the solid fraction using a laser particle size analyzer. If it deviates from the target range, adjust the heating power or stirring speed.

5. The real-time control method for the semi-solid forming process of the universal joint fork arm according to claim 1, characterized in that, The process of quantitatively conveying the prepared semi-solid slurry to the mold cavity, adjusting the heating power of the heating coil wound around the outer wall of the conveying pipe in real time using a PID temperature controller, and installing an infrared temperature sensor at the outlet of the conveying pipe to monitor the slurry outlet temperature in real time specifically includes: Based on the volume of the universal joint fork arm, the required slurry volume is calculated, and the matching parameters of the conveying flow rate, conveying speed and screw speed of the servo screw conveyor are set. An infrared temperature sensor is installed at the outlet of the conveying pipeline, and the angle of the sensor is adjusted to face the slurry outlet to capture the slurry temperature in real time. During the semi-solid slurry transportation process, the pipeline temperature is monitored in real time by a PID controller. Combined with the outlet temperature feedback from the infrared temperature probe, if the slurry temperature loss exceeds 5°C or the outlet temperature is lower than the lower limit of the target range, the power of the heating coil is increased. If the outlet temperature is higher than the upper limit of the target range, the power of the heating coil is reduced.

6. The real-time control method for the semi-solid forming process of the universal joint fork arm according to claim 1, characterized in that, The process involves starting the hydraulic press to close the mold and apply pressure. Temperature, pressure, and temperature field distribution within the mold cavity are collected via thermocouples, pressure sensors, and an infrared thermal imager. Based on a BP neural network-based forming parameter prediction model, predicted forming density and dimensional accuracy data for the universal joint fork arm are obtained. Multi-field coupling control based on thresholds is then implemented, specifically including: Start the hydraulic press to drive the upper mold to move downwards and close the mold. When the mold is closed to the preset position, switch from the mold closing mode to the pressurization mode. The temperature and pressure data of the fork head, rod, and fork root are collected in real time by K-type thermocouples and pressure sensors installed in the cavity, and the overall temperature field distribution data of the cavity is captured by an infrared thermal imager. By inputting the collected data into the pre-trained BP neural network model, the forming density and dimensional accuracy data of the universal joint fork arm are obtained. Based on the difference between the predicted data and the preset threshold, the forming process is controlled by adjusting the power of the auxiliary heating device or controlling the pressure of the hydraulic press.

7. The real-time control method for the semi-solid forming process of the universal joint fork arm according to claim 1, characterized in that, Once the slurry fills the mold cavity, the pressure holding stage begins. Displacement sensors monitor the displacement changes of the hydraulic press slider in real time to indirectly determine the solidification shrinkage of the blank. If the rate of temperature drop in the local area of ​​the cavity exceeds 2℃ / s, the auxiliary heating device for the corresponding area of ​​the mold is activated for pressure compensation. Specific details include: When the mold cavity pressure reaches the preset threshold, the system switches from pressurization mode to pressure holding mode, monitors displacement sensor data, and if the slider displacement changes by more than 0.1mm, increases the pressure holding pressure and extends the pressure holding time to compensate for shrinkage. The temperature drop rate of each area is monitored in real time by a type K thermocouple. If the temperature drop rate of a certain area exceeds 2℃ / s, the ceramic heating element in that area is activated to slow down the temperature drop. After adjustment, continuously monitor pressure, displacement, and temperature data until the pressure holding time ends.

8. The real-time control method for the semi-solid forming process of the universal joint fork arm according to claim 1, characterized in that, After the pressure holding period, demolding is performed. A coordinate measuring machine is used to detect the key dimensions of the universal joint fork arm in real time. An ultrasonic flaw detector is used to detect internal defects in the workpiece. The temperature range, pressure, or pressure holding time parameters for the next forming process are then adjusted. Specifically, this includes: After the pressure holding stage is completed, switch to demolding mode and observe the fit between the workpiece and the cavity. If slight adhesion occurs, fine-tune the demolding speed. After demolding, the universal joint fork arm is fixed on the worktable of the coordinate measuring machine. Key dimensions such as fork spacing, rod diameter, and head thickness are selected for inspection, and the deviations between the actual values ​​and the design values ​​of each dimension are recorded. The head and root area of ​​the workpiece are scanned by an ultrasonic flaw detector to detect whether there are shrinkage cavities or shrinkage porosity inside, and the location and size of the defects are marked. Based on dimensional deviations exceeding thresholds or the presence of defects exceeding standards, the temperature range, pressure, and holding time for the next forming process are adjusted to form a closed-loop control.

9. A real-time control system for the semi-solid forming process of a universal joint fork arm, used to implement the real-time control method for the semi-solid forming process of a universal joint fork arm as described in any one of claims 1-8, characterized in that, include: Mold condition monitoring module: By polishing the mold cavity and installing thermocouples, pressure sensors and infrared thermal imagers, it can collect real-time data on cavity temperature, pressure and temperature field distribution. Slurry state control module: Controls slurry viscosity and solid fraction through medium-frequency induction heating and stirring, and dynamically adjusts heating power or stirring speed to maintain target parameters; Forming process prediction and control module: Based on the BP neural network model, it predicts the forming density and dimensional accuracy in real time, and optimizes the process parameters through multi-field coupling control; Pressure holding compensation module: Monitors slider displacement and local temperature changes, dynamically adjusts pressure holding pressure or activates auxiliary heating device to compensate for contraction and sudden temperature drop; Quality inspection feedback module: By detecting the size and internal defects of the workpiece, the feedback data is used to correct the forming parameters for the next time, realizing automatic iterative optimization of process parameters and closed-loop control of the whole process; Processor: The processor is used to handle the calculation process of each formula and the construction calculation process of each model.

Images