Nozzle pressure measurement and control system capable of measuring ultrahigh pressure of 4000 bar for multiple times

Abstract

The invention discloses a nozzle pressure measurement and control system and method capable of measuring ultrahigh pressure of 4000 bar for multiple times, and the system comprises a pressure sensing module which employs a GRAEFF microelectronic technology, is equipped with an active cooling and heat insulation structure, and can stably work in an ultrahigh-pressure and high-temperature environment; the signal processing module is integrated with a temperature compensation algorithm based on a BP neural network, so that the signal precision is remarkably improved; the core control module is combined with a self-adaptive fuzzy PID controller and a feedforward compensator to form an intelligent composite control strategy; and the high-response executing mechanism accurately drives the injection action. The method comprises the steps of real-time pressure acquisition and intelligent compensation, pressure deviation calculation, self-adaptive fuzzy PID and feedforward compound control quantity calculation, and comprehensive output and pressure adjustment. According to the invention, the key technical problems of large pressure measurement drift, slow control response and easy oscillation in ultrahigh pressure injection molding are solved, long-term high-precision measurement and rapid and accurate control under the extreme working condition of 4000 bar / 300 DEG C are realized, and the quality consistency and qualified rate of precision injection molding products are obviously improved.

Description

Technical Field

[0001] This invention relates to the field of plastic injection molding technology, and in particular to a nozzle pressure measurement and control system capable of repeatedly measuring ultra-high pressure of 4000 bar. Specifically, it relates to a nozzle pressure measurement and control system for ultra-high pressure injection molding machines, which is particularly suitable for long-term stable, high-precision, and repetitive pressure measurement and rapid closed-loop control under extreme working conditions with pressures as high as 4000 bar (approximately 400 MPa) and above. Background Technology

[0002] With the rapid development of precision engineering and micro / nano manufacturing technologies, injection molding technology is constantly advancing towards higher precision and performance. When molding high-end products such as micro / nano structure parts, optical lenses, precision gears, and medical device components, extremely stringent requirements are placed on melt filling and replication fidelity. To overcome the enormous flow resistance encountered by the melt during rapid filling in high aspect ratio microchannels and to ensure that the material can accurately replicate every minute feature of the mold cavity, extremely high injection pressures are necessary. Therefore, ultra-high pressure injection molding technology with injection pressures reaching 4000 bar (approximately 400 MPa) and above has become a key development direction in this field.

[0003] However, the ultra-high pressure environment presents unprecedented challenges to injection molding machines, especially to their core component—the pressure measurement and control system at the nozzle. These challenges are mainly reflected in the following aspects: Firstly, in pressure measurement, traditional piezoelectric or strain gauge pressure sensors, when subjected to ultra-high pressures of 4000 bar and molten temperatures exceeding 300°C, experience significant changes in the physical properties of their core sensing elements. This leads to severe drift in the sensor's zero-point output and sensitivity. Zero-point drift causes inaccuracy of the static pressure reference, while sensitivity drift results in distortion of dynamic pressure measurements. Prolonged operation under these extreme conditions not only drastically reduces measurement accuracy but also significantly shortens the sensor's lifespan, potentially causing rapid failure due to thermal fatigue or overload. While existing technologies have attempted to mitigate this problem through hardware insulation or simple linear temperature compensation, their effectiveness is limited under the extreme coupling conditions of 4000 bar / 300°C, failing to meet the requirements for long-term stable and high-precision measurements.

[0004] Secondly, regarding pressure control, ultra-high pressure injection molding is a typical nonlinear, time-varying dynamic system. Under ultra-high pressure, the viscosity of the polymer melt is extremely sensitive to minute fluctuations in pressure and temperature. Coupled with instantaneous changes in injection speed, this can easily trigger violent oscillations in nozzle pressure. Conventional proportional-integral-derivative (PID) control systems, whose parameters are usually based on linearized models or tuned at specific operating points, struggle to adapt to such drastic parameter changes and nonlinear characteristics. Therefore, conventional PID controllers have relatively slow response speeds and insufficient dynamic adjustment capabilities, easily leading to significant pressure overshoot or continuous oscillations. This unstable pressure control directly impacts product quality, causing various defects such as flash, short runs, shrinkage marks, and excessive internal stress, severely affecting the stability of the ultra-high pressure injection molding process and the product yield.

[0005] In summary, existing technologies lack a comprehensive solution capable of achieving long-term stable and high-precision measurement under ultra-high pressure (4000 bar) and continuous high-temperature conditions, while simultaneously possessing rapid response and strong anti-interference capabilities to achieve accurate, multi-stage pressure closed-loop control. This has become a technical bottleneck restricting the further development and widespread application of ultra-high pressure injection molding technology. Summary of the Invention

[0006] To address the shortcomings of existing technologies, the purpose of this invention is to provide a nozzle pressure measurement and control system capable of repeatedly measuring ultra-high pressure of 4000 bar. This system can perform long-term, stable, and high-precision measurement of melt pressure at the injection molding machine nozzle under pressures of 4000 bar and above and high temperatures exceeding 300°C, and achieve rapid and accurate closed-loop control to meet the process requirements of multiple injection cycles.

[0007] Another objective of this invention is to provide a pressure control method based on the above system, which can effectively suppress pressure oscillations under ultra-high pressure, quickly track the set pressure curve, ensure the stability of the molding process, and improve the consistency of product quality.

[0008] The above-mentioned objective of this invention is achieved through the following technical solutions: On one hand, this invention discloses a nozzle pressure measurement and control system capable of repeatedly measuring ultra-high pressure of 4000 bar, comprising: The pressure sensing module is used to measure the melt pressure at the nozzle of the injection molding machine in real time; The signal processing module is electrically connected to the pressure sensing module and is used to condition and temperature compensate the raw signal output by the pressure sensing module. The core control module, which is communicatively connected to the signal processing module, is used to receive the processed actual pressure value, compare it with the set pressure value, and calculate the control signal based on the deviation. The actuator is connected to the core control module and is used to adjust the injection action of the injection molding machine according to the control signal in order to control the melt pressure at the nozzle. The pressure sensing module includes an ultra-high temperature and high pressure pressure sensor based on GRAEFF microelectronics technology.

[0009] As a further technical solution of the present invention: the pressure sensing module further includes: An active cooling structure, which is a double-layer water-cooling jacket surrounding the base of the pressure sensor, is used to introduce constant-temperature cooling water to actively cool the sensor. Passive thermal insulation structure, including high-temperature thermal insulation material filled at the sensor signal leads.

[0010] As a further technical solution of the present invention: the signal processing module includes a temperature compensation unit, which uses a neural network-based software algorithm to compensate for the zero-point drift and sensitivity drift of the pressure sensor in real time.

[0011] Through the above technical solution, the temperature compensation unit is used to receive the original output signal of the pressure sensor and the reading of its internal temperature sensor, and to perform real-time nonlinear temperature compensation on the original output signal using a software compensation algorithm based on a BP neural network, and output a high-precision actual pressure value after compensation; the signal processing module also includes a signal conditioning circuit, which is used to amplify and filter the weak electrical signal output by the pressure sensor, and convert it into a standard industrial control signal.

[0012] As a further technical solution of the present invention: the core control module includes: An adaptive fuzzy PID controller is used to adjust PID parameters online based on pressure deviation and the rate of change of deviation. A feedforward compensator is used to generate a feedforward control signal based on the injection speed setpoint. The output of the feedforward compensator is superimposed with the output of the adaptive fuzzy PID controller to form a control signal for the actuator.

[0013] Through the above technical solution, the core control module also includes a pressure setting unit, which is used to store and provide the expected pressure curve P_set(t) for each stage of the injection process; An adaptive fuzzy PID controller is used to receive the actual pressure value P_actual output by the signal processing module, calculate the pressure deviation e(n) = P_set - P_actual and the deviation change rate ec(n) = e(n) - e(n-1), and adjust the proportional coefficient Kp, integral coefficient Ki and derivative coefficient Kd of the PID controller online using fuzzy inference rules, thereby calculating the PID control quantity u_pid(n). The feedforward compensator is used to generate a feedforward control quantity u_ff(n) based on a preset injection speed setting curve. The feedforward control quantity u_ff(n) and the PID control quantity u_pid(n) are superimposed to form a comprehensive control signal u(n) for the actuator.

[0014] As a further technical solution of the present invention: In the adaptive fuzzy PID controller, the input variables of fuzzy inference are pressure deviation e and its rate of change ec, and the output variables are the correction amounts ΔKp, ΔKi and ΔKd of the PID parameters, where ΔKp is the increment of the proportional coefficient, ΔKi is the increment of the integral coefficient and ΔKd is the increment of the derivative coefficient.

[0015] Through the above technical solution, the fuzzy inference rule is as follows: when the absolute value of the pressure deviation |e| is large, the parameter is adjusted to enhance the proportional effect and weaken the derivative effect; when the absolute value of the pressure deviation |e| is small, the parameter is adjusted to weaken the proportional effect and enhance the integral effect.

[0016] As a further technical solution of the present invention: the actuator is a high-response electro-hydraulic servo valve with a frequency response of not less than 100Hz and its matching servo driver, used to control the injection cylinder of a hydraulic injection molding machine.

[0017] As a further technical solution of the present invention: the actuator is a high dynamic response servo motor and a precision ball screw drive system, used to control the screw movement of an all-electric injection molding machine.

[0018] On the other hand, the present invention also discloses a method for controlling the nozzle pressure of an ultra-high pressure injection molding machine applied to the above-mentioned system, comprising the following steps: S1: System initialization, load the expected pressure curve, and initialize controller parameters; S2: Real-time acquisition of nozzle pressure raw signal and sensor temperature signal, and temperature compensation using a pre-trained BP neural network model to obtain a high-precision actual pressure value P_actual; S3: Calculate the pressure deviation e(n) = P_set(n) - P_actual(n) and the deviation change rate ec(n) = e(n) - e(n-1) at the current moment; S4: Using an adaptive fuzzy PID controller, the PID parameters are adjusted online based on e(n) and ec(n), and the PID control quantity u_pid(n) is calculated; at the same time, the feedforward compensator calculates the feedforward control quantity u_ff(n) based on the injection speed setpoint. S5: Add the PID control quantity u_pid(n) and the feedforward control quantity u_ff(n) to obtain the comprehensive control signal u(n); S6: Drive the actuator according to the comprehensive control signal u(n) to adjust the injection speed or pressure so that the actual melt pressure at the nozzle tracks the set pressure curve; S7: Repeat steps S2 to S6 during the injection and pressure holding phases until the process is complete.

[0019] As a further technical solution of the present invention: In step S2, the BP neural network compensation model is a three-layer feedforward network. Its input is the original output voltage value of the pressure sensor and the reading of the internal temperature sensor, and its output is the accurate pressure value after nonlinear temperature compensation. The model is obtained by training with a large amount of calibration data covering the pressure range of 0-5000 bar and the temperature range of 30-320°C.

[0020] As a further technical solution of the present invention: In step S4, the specific process of adjusting the PID parameters online includes: The precise values ​​of pressure deviation e and deviation change rate ec are fuzzified to obtain the corresponding fuzzy subsets; Fuzzy inference is performed based on a preset fuzzy rule base to obtain fuzzy subsets of output variables ΔKp, ΔKi, and ΔKd; Defuzzify the output fuzzy subset to obtain accurate PID parameter correction values; Update the parameters of the current PID controller using the correction amount: Kp = Kp_prev + ΔKp, Ki = Ki_prev + ΔKi, Kd = Kd_prev + ΔKd.

[0021] In summary, compared with the prior art, the present invention has at least one of the following beneficial technical effects: High measurement accuracy, excellent long-term stability, and long lifespan: Utilizing a dedicated high-voltage sensor based on GRAEFF microelectronics technology, its materials and structure are optimized for high-temperature and high-pressure environments, possessing inherent high stability. An innovative dual temperature drift suppression strategy of "hardware insulation + software compensation": Hardware measures such as double-layer water-cooling jackets and high-temperature insulation cotton significantly reduce the operating temperature of the sensor core; combined with a nonlinear software compensation algorithm based on BP neural networks, residual and inherent temperature drift of the sensor are accurately modeled and corrected. The synergistic effect of these two measures greatly suppresses zero-point drift and sensitivity drift under extreme conditions of 4000 bar / 300°C, achieving high-precision measurement over long periods (e.g., more than thousands of injection cycles) (e.g., full-range accuracy better than ±0.5% FS). The sensor's ultra-hard wear-resistant coating and robust structure ensure its durability under high-pressure molten metal impact.

[0022] The system boasts fast response, excellent dynamic performance, and minimal overshoot: it employs a composite intelligent control strategy of "adaptive fuzzy PID + feedforward compensation." Adaptive fuzzy PID can identify the system's dynamic characteristics online and adjust parameters, effectively overcoming the nonlinearity and time-varying nature of the ultra-high pressure injection molding process. It offers faster response and stronger adaptability than traditional PID. Feedforward compensation proactively adjusts for the main measurable disturbances (injection speed changes), complementing feedback control and significantly reducing pressure shocks and oscillations caused by sudden speed changes. This composite control strategy together ensures the system has extremely fast tracking capability and strong anti-interference ability to the set pressure curve, effectively avoiding pressure overshoot and continuous oscillations, resulting in a smooth pressure build-up process and a small steady-state fluctuation range (e.g., near the 4000 bar setpoint, fluctuations can be controlled within ±20 bar).

[0023] The system boasts strong robustness and high consistency in molding quality: Through a multi-layered, systematic design encompassing the sensing layer (thermal management), signal layer (intelligent compensation), and control layer (intelligent control), the entire system maintains extremely high stability and reliability even under complex conditions such as ultra-high pressure, high temperature, and fluctuations in melt characteristics. Stable and precise pressure control is crucial for ensuring product quality. This invention effectively solves the core problems of inaccurate pressure measurement and unstable control in ultra-high pressure injection molding, thereby significantly reducing defects such as flash, short material, and internal stress caused by pressure issues, and greatly improving the consistency of molding quality and the production pass rate of precision injection molded products. Attached Figure Description

[0024] Figure 1 This is a block diagram of the overall structure of the system of the present invention.

[0025] Figure 2 This is a schematic diagram of the pressure sensing module structure of the present invention.

[0026] Figure 3 This is a schematic diagram of the installation structure of the pressure sensing module and nozzle of the present invention.

[0027] Figure 4 To and Figure 3 A schematic diagram of the structure inside the nozzle and sensor pressure-sensing chamber.

[0028] Figure 5 This is a diagram showing the nozzle pressure curve of the present invention.

[0029] Reference numerals: 1. Pressure sensing module; 11. Nozzle connector; 12. Pressure sensor; 121. Sensor cable; 13. Heat insulation sheet; 14. Magnetic connector; 15. Magnetic base; 2. Core control module; 3. Mold; 31. Mounting base; 32. Cavity; 33. Barrel; 34. Screw; 35. Heating coil; 351. Heating coil cable. Detailed Implementation

[0030] The technical solutions in the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. All other embodiments obtained by those skilled in the art based on the embodiments of this application without creative effort are within the scope of protection of this application.

[0031] In the description of this application, it should be noted that the terms "upper," "lower," "inner," "outer," "top / bottom," etc., indicating the orientation or positional relationship are based on the orientation or positional relationship shown in the accompanying drawings, and are only for the convenience of describing this application and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation, and therefore should not be construed as a limitation of this application. Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance.

[0032] In the description of this application, it should be noted that, unless otherwise expressly specified and limited, the terms "installed," "equipped with," "sleeved / connected," "connected," etc., should be interpreted broadly. For example, "connection" can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection or an indirect connection through an intermediate medium; it can be a connection within two components. Those skilled in the art can understand the specific meaning of the above terms in this application based on the specific circumstances.

[0033] Example 1: Reference Figure 1 This invention discloses a nozzle pressure measurement and control system capable of repeatedly measuring ultra-high pressure of 4000 bar, comprising: a pressure sensing module 1 for real-time measurement of melt pressure at the nozzle of an injection molding machine; a signal processing module electrically connected to the pressure sensing module 1 for conditioning and temperature compensation of the raw signal output by the pressure sensing module 1; a core control module 2 communicatively connected to the signal processing module for receiving the processed actual pressure value, comparing it with a set pressure value, and calculating a control signal based on the deviation; and an actuator connected to the core control module 2 for adjusting the injection action of the injection molding machine according to the control signal to control the melt pressure at the nozzle.

[0034] The pressure sensing module 1 includes an ultra-high temperature and high pressure pressure sensor 12 based on GRAEFF microelectronics technology. The core sensing element is the GRAEFF microelectronics-based pressure sensor 12. This sensor employs special magnetic materials and a thermal insulation structure design to effectively isolate its internal sensing element from the external high-temperature environment. Its pressure-sensing diaphragm surface is deposited with an ultra-hard, wear-resistant, and insulating coating (such as a diamond-like carbon coating or an alumina ceramic coating). This coating not only possesses extremely high hardness and wear resistance, capable of withstanding high-speed erosion by the melt, but also exhibits excellent insulation properties, preventing electrochemical corrosion. The sensor is directly and tightly installed in the pre-reserved sensing hole on the side wall or end of the injection molding machine nozzle body via a high-strength alloy steel shell and its threaded structure, ensuring direct contact between its pressure-sensing surface and the melt flow channel. This direct-insertion installation method ensures the authenticity and speed of pressure transmission. The GRAEFF pressure sensor 12 can directly withstand instantaneous pressures up to 5000 bar and continuous melt temperatures above 300°C. Under the action of the heating coil 35, the melt in the nozzle and sensor pressure chamber always remains in a liquid state, which provides a physical basis for continuous and multiple pressure measurements at different injection stages (injection, holding pressure).

[0035] The pressure sensing module 1 also includes: an active cooling structure, which is a double-layer water-cooled jacket surrounding the base of the pressure sensor 12, used to actively cool the sensor by circulating constant-temperature cooling water; and a passive thermal insulation structure, including high-temperature thermal insulation material filled at the sensor signal leads. A double-layer water-cooled jacket is provided in the assembly area between the sensor base and the injection molding machine nozzle mounting hole. This jacket surrounds the sensor base and is circulated with constant-temperature cooling water (e.g., temperature controlled between 20°C and 50°C) provided by an external cooling system. The circulating cooling water continuously removes a large amount of heat conducted from the nozzle to the sensor mounting area, forming a localized low-temperature zone to ensure that the sensor base temperature is maintained within a safe range (typically below 100°C). At the sensor signal lead outlet, a high-performance high-temperature thermal insulation material, such as ceramic wool or aerogel felt, is filled. This insulation material forms an effective thermal barrier, preventing high temperatures from being conducted along the leads to the signal processing unit and protecting the electronic circuitry from heat damage.

[0036] like Figure 2 and Figure 3As shown, the pressure sensing module 1 includes, from top to bottom, a nozzle connector 11, a pressure sensor 12, a heat insulation sheet 13, a magnetic connector 14, and a magnetic base 15. The magnetic base 15 is mounted on the mounting base 31 of the mold 3 and located on one side of the cavity 32. The nozzle connector 11 is connected to the discharge end of the material cylinder 33. A screw 34 is installed inside the material cylinder 33. A sensor cable 121 is connected to the pressure sensor 12. A heating coil 35 is installed on the outer wall of the pressure sensor 12, and a temperature sensor is integrated inside the heating coil 35. A heating coil 35 cable is connected to the heating coil 35. Figure 4 As shown, within the nozzle and sensor pressure chamber, the melt pressure and the force acting on the sensor are in the same direction.

[0037] The signal processing module is connected to the output of the pressure sensing module 1 and is responsible for preprocessing and compensating the raw signal output by the sensor to obtain a high-precision actual pressure value.

[0038] The signal processing module includes a temperature compensation unit, which uses a neural network-based software algorithm to compensate for the zero-point drift and sensitivity drift of the pressure sensor 12 in real time. The temperature compensation unit receives the raw output signal from the pressure sensor 12 and the readings from its internal temperature sensor, and uses a BP neural network-based software compensation algorithm to perform real-time nonlinear temperature compensation on the raw output signal, outputting a high-precision actual pressure value after compensation.

[0039] The temperature compensation unit is crucial for improving measurement accuracy. The GRAEFF sensor integrates a high-precision platinum resistance or semiconductor temperature sensor for real-time monitoring of the sensor chip's junction temperature. The temperature compensation unit employs a software compensation algorithm based on a back propagation (BP) neural network. This algorithm model was trained offline using extensive calibration experimental data covering various pressure (0-5000 bar) and temperature (room temperature to 350°C) combinations before system deployment. The trained model is stored in the signal processing module's memory. During operation, the compensation unit simultaneously reads the raw output signal (typically millivolt-level voltage) from the pressure sensor 12 and the reading from the internal temperature sensor, using these as inputs to the BP neural network model. After forward calculation, the model outputs the accurate pressure value in real-time after nonlinear temperature compensation. This compensation method effectively corrects zero-point drift and sensitivity drift caused by temperature.

[0040] The signal processing module also includes a signal conditioning circuit, used to amplify and filter the weak electrical signal output by the pressure sensor 12, and convert it into a standard industrial control signal. This circuit is responsible for hardware-level preprocessing of the weak analog signal output by the sensor. It typically includes: Amplification circuit: A low-noise, high-input-impedance instrumentation amplifier is used to amplify the millivolt-level differential voltage signal to a suitable range for acquisition (e.g., 0-5V or 0-10V).

[0041] Filtering circuit: Use a low-pass filter (such as a Butterworth filter or a Chebyshev filter) to filter out high-frequency noise (such as electromagnetic interference and vibration noise) in the signal and improve the signal-to-noise ratio.

[0042] Signal conversion and isolation: The amplified and filtered voltage signal is converted into a standard industrial control signal, such as a 4-20mA current signal or a 0-10V voltage signal, to enhance anti-interference capability and transmission distance. Optionally, opto-isolation or magnetic isolation devices can be added to the circuit to achieve electrical isolation between the signal processing module and the subsequent control module, preventing ground loop interference.

[0043] The core control module 2 is the brain of the system. It receives high-precision pressure signals from the signal processing module and generates control commands according to the control strategy.

[0044] Core control module 2 includes: an adaptive fuzzy PID controller, used to adjust PID parameters online according to pressure deviation and deviation change rate; a feedforward compensator, used to generate a feedforward control signal according to the injection speed setpoint; the output of the feedforward compensator and the output of the adaptive fuzzy PID controller are superimposed to form the control signal for the actuator.

[0045] The core control module 2 also includes a pressure setting unit, used to store and provide the expected pressure curves P_set(t) for each stage of the injection process; the pressure setting unit is used to store and provide the expected pressure curves for each stage of the injection process (injection stage, holding pressure stage). This curve can be a step curve, a ramp curve, or a more complex multi-segment curve, which is preset by the process engineer according to the specific product and material characteristics.

[0046] An adaptive fuzzy PID controller receives the actual pressure value P_actual from the signal processing module, calculates the pressure deviation e(n) = P_set - P_actual and the rate of change of deviation ec(n) = e(n) - e(n-1), and uses fuzzy inference rules to adjust the proportional coefficient Kp, integral coefficient Ki, and derivative coefficient Kd of the PID controller online, thereby calculating the PID control quantity u_pid(n). The adaptive fuzzy PID controller is an intelligent controller that combines the stability of traditional PID control with the flexibility of fuzzy logic to handle nonlinearities. Its working principle is as follows: The system receives the real-time, compensated actual pressure value P_actual from the signal processing module and compares it with the set pressure value P_set provided by the pressure setting unit to calculate the current control deviation e(n) = P_set - P_actual.

[0047] At the same time, the rate of change of the deviation is calculated as ec(n) = e(n) - e(n-1), which reflects the trend and speed of pressure change.

[0048] The deviation e(n) and the rate of change of deviation ec(n) are used as the two input variables of the fuzzy inference system. The fuzzy inference system has a pre-stored fuzzy rule base, for example: "If the deviation is large positive and the rate of change of deviation is small negative, then the increment of the output proportional coefficient is large positive." Through fuzzification, rule inference, and defuzzification steps, the adjustment amounts (ΔKp, ΔKi, ΔKd) of the three parameters (proportional coefficient Kp, integral coefficient Ki, and derivative coefficient Kd) of the traditional PID controller are calculated in real time online.

[0049] These adjustments are used to correct the parameters of the PID controller online. The corrected PID controller then calculates the PID control quantity u_pid(n) based on the current e(n) and ec(n). This self-tuning mechanism enables the controller to adapt to changes in the system's dynamic characteristics caused by variations in melt viscosity, temperature, etc., during the injection process.

[0050] A feedforward compensator generates a feedforward control quantity u_ff(n) based on a preset injection speed setpoint curve. This feedforward control quantity u_ff(n) is superimposed with the PID control quantity u_pid(n) to form a comprehensive control signal u(n) for the actuator. To further improve the system's response speed and anti-interference capability, especially to suppress pressure disturbances caused by sudden changes in the injection speed setpoint, feedforward control is introduced. The feedforward compensator directly generates a feedforward control signal u_ff(n) based on a pre-set and known injection speed curve (usually a function of screw displacement or speed over time) through a pre-established lookup table or functional relationship. This signal does not depend on pressure feedback deviation but rather acts "ahead of time" to counteract predictable disturbances. The output of the feedforward compensator is algebraically added to the output of the adaptive fuzzy PID controller at the superposition point.

[0051] In an adaptive fuzzy PID controller, the input variables for fuzzy inference are the pressure deviation *e* and its rate of change *ec*, and the output variables are the correction values ​​ΔKp, ΔKi, and ΔKd for the PID parameters. ΔKp is the increment of the proportional gain, ΔKi is the increment of the integral gain, and ΔKd is the increment of the derivative gain. The fuzzy inference rule is: when the absolute value of the pressure deviation |e| is large, adjusting the parameters strengthens the proportional gain and weakens the derivative gain; when the absolute value of the pressure deviation |e| is small, adjusting the parameters weakens the proportional gain and strengthens the integral gain.

[0052] The actuator consists of a high-response electro-hydraulic servo valve with a frequency response of not less than 100Hz and its matching servo driver, used to control the injection cylinder of the hydraulic injection molding machine. The actuator also consists of a high-dynamic-response servo motor and a precision ball screw drive system, used to control the screw 34 movement of the all-electric injection molding machine.

[0053] The actuator receives the integrated control signal from the core control module 2 and converts it into physical actions, ultimately adjusting the pressure and speed of the injection cylinder to control the melt pressure at the nozzle.

[0054] For hydraulic injection molding machines: the actuator is mainly a high-response servo valve. This servo valve is typically a two- or three-stage electro-hydraulic servo valve with a frequency response of not less than 100Hz. It receives analog voltage or current control signals output from the core control module 2, and precisely and quickly adjusts its valve core opening, thereby controlling the flow and pressure of hydraulic oil entering the injection cylinder, driving the screw 34 forward, and achieving precise control of injection speed and pressure.

[0055] For all-electric injection molding machines: the actuators mainly consist of high-dynamic-response servo motors and their drivers, as well as precision ball screw pairs. The servo driver receives control signals, controls the speed and torque of the servo motor, and converts the rotary motion into the linear motion of the screw 34 through the ball screw, thereby achieving precise pressure control.

[0056] In this embodiment, the pressure sensing module 1 is tightly screwed into a specially machined high-strength sensing mounting hole on the side wall of the injection molding machine nozzle via its external high-strength threads. During installation, ensure that the sensor pressure-sensing diaphragm is flush with the inner wall of the melt flow channel to reduce flow dead zones and material accumulation. The sensor selected in this embodiment is a GRAEFF ultra-high temperature and high pressure pressure sensor 12 with a range of 5000 bar. A double-layer stainless steel water-cooling jacket is installed between the sensor base and the nozzle mounting hole wall. The cooling water inlet and outlet are connected to an external constant temperature circulating cooling water machine, with the water supply temperature set at 25±1°C and the flow rate at 5L / min. After the sensor signal lead is led out from the mounting base 31, the gaps around it are tightly filled with alumina ceramic cotton as a heat insulation layer. A copper alloy heat insulation sheet 13 can also be installed between the sensor housing and the mounting base 31 to further block heat conduction.

[0057] The signal processing module can be integrated into the sensor junction box or be a standalone control cabinet module. Its core is a circuit board with an analog-to-digital converter (ADC) and a microprocessor (MCU). The onboard memory pre-stores a trained BP neural network compensation model for this specific sensor. The model's input layer has two nodes: the sensor's raw voltage output (the digital value converted by the ADC) and the resistance value of the integrated temperature sensor (the converted digital temperature value). The hidden layer is designed with 5 neurons, and the output layer has 1 node, representing the compensated, precise pressure value. During offline training, the model used thousands of data sets from a pressure calibration machine, covering a full range of combinations from 0 to 5000 bar (in 500-bar steps) and from 30°C to 320°C (in 20-°C steps).

[0058] The core control module 2 is implemented using an industrial-grade programmable automation controller (PAC) or a high-end PLC (such as the Siemens S7-1500 series). Its internal control program performs the following functions: Pressure setting unit: Stores the injection-holding pressure curve for a certain precision optical lens. During the injection phase, the pressure linearly increases from 0 to 4000 bar within 0.1 seconds, and during the holding phase, it is maintained at 3800 bar and 3500 bar in two segments within 2 seconds.

[0059] Adaptive fuzzy PID controller: Its fuzzy rule table has the following example rules as preset: If |e| is large (positive or negative) and |ec| is small, then take the larger ΔKp (positive) and the smaller ΔKd (negative) to quickly eliminate the main deviation.

[0060] If |e| is moderate and ec has the same sign as e (the deviation is widening), then take a moderate ΔKp (positive) and a smaller ΔKi (negative) to suppress the widening of the deviation.

[0061] If |e| is small and |ec| is small, then we take a smaller ΔKp (negative small) and a larger ΔKi (positive large) to reduce the steady-state error.

[0062] The universes of discourse for input variables e and ec are both defined as [-500, 500] (bar), and are fuzzified using fuzzy subsets (negative large, negative medium, negative small, zero, positive small, positive medium, positive large). The universes of discourse for output variables ΔKp, ΔKi, and ΔKd are set empirically, and precise correction values ​​are obtained through defuzzification (such as the centroid method).

[0063] Feedforward compensator: Its lookup table is determined experimentally during the system debugging phase based on the screw diameter (34mm), injection cylinder area, and rheological properties of commonly used materials of the injection molding machine. For example, when the injection speed command increases from 0 to 80% of the maximum value within 0.1 seconds, a positive feedforward signal u_ff with an amplitude of approximately 30% of the total control quantity is output.

[0064] The actuator is a high-performance two-stage electro-hydraulic servo valve with a frequency response of 120Hz, such as the Moog D-series servo valve. This servo valve receives a ±10V analog voltage control signal u(n) from the PAC via a servo driver, precisely controlling its valve core displacement, thereby regulating the hydraulic oil flow to the injection cylinder, driving the screw 34 forward, and ultimately achieving precise control of the melt pressure at the nozzle.

[0065] Example 2: On the other hand, the present invention also discloses a method for controlling the nozzle pressure of an ultra-high pressure injection molding machine applied to the system of the above embodiment one, comprising the following steps: S1: System initialization, loading the expected pressure curve, and initializing controller parameters; specifically: starting the system, loading the expected pressure curve P_set(t) for the injection and holding phases into core control module 2. Initializing the initial parameters (Kp0, Ki0, Kd0) and fuzzy rule base of the adaptive fuzzy PID controller. Initializing the lookup table of the feedforward compensator; S2: Real-time acquisition of the nozzle pressure raw signal and sensor temperature signal, and temperature compensation using a pre-trained BP neural network model to obtain a high-precision actual pressure value P_actual; specifically: during the injection process, the pressure sensing module 1 acquires the melt pressure at the nozzle in real time and outputs the raw electrical signal. The signal processing module simultaneously acquires the temperature signal inside the pressure sensor 12. Using the pre-trained BP neural network compensation model, real-time, online temperature compensation is performed on the raw pressure signal to obtain a high-precision actual pressure value P_actual; S3: Core control module 2 reads the current set pressure value P_set(n) and the compensated actual pressure value P_actual(n); calculates the current pressure deviation e(n) = P_set(n) - P_actual(n) and the deviation change rate ec(n) = e(n) - e(n-1); S4: Using an adaptive fuzzy PID controller, the PID parameters are adjusted online based on e(n) and ec(n), and the PID control quantity u_pid(n) is calculated. Simultaneously, the feedforward compensator calculates the feedforward control quantity u_ff(n) based on the injection speed setpoint. Specifically, the adaptive fuzzy PID controller fuzzifies e(n) and ec(n), and through a fuzzy inference rule base, adjusts and obtains the correction values ​​ΔKp, ΔKi, and ΔKd of the PID parameters online. Then, using the adjusted parameters (Kp = Kp0 + ΔKp, Ki = Ki0 + ΔKi, Kd = Kd0 + ΔKd), the PID control quantity u_pid(n) is calculated. Simultaneously, the feedforward compensator obtains the feedforward control quantity u_ff(n) based on the current injection speed setpoint through table lookup or calculation.

[0066] S5: Add the PID control quantity u_pid(n) and the feedforward control quantity u_ff(n) to obtain the comprehensive control signal u(n); superimpose the u_pid(n) calculated by the adaptive fuzzy PID controller and the u_ff(n) calculated by the feedforward compensator to obtain the final comprehensive control quantity u(n) = u_pid(n) + u_ff(n). This comprehensive control signal is output to the actuator; S6: Drive the actuator according to the integrated control signal u(n) to adjust the injection speed or pressure, so that the actual melt pressure at the nozzle tracks the set pressure curve; specifically: the actuator (high-response servo valve or servo motor) receives the control signal u(n) and adjusts its opening or rotation speed according to the magnitude and polarity of the signal. This changes the advance speed or torque of the injection cylinder, thereby dynamically adjusting the extrusion force of the screw 34 on the melt, ultimately causing the melt pressure P_actual at the nozzle to change towards the set pressure P_set. S7: During the injection and holding pressure stages, steps S2 to S6 are executed cyclically until the process ends. During the injection and holding pressure stages, the system continuously repeats steps 2 to 6 at extremely high sampling and control frequencies (e.g., above 1kHz) to collect, compensate, calculate, and output data in real time, forming a fast and accurate closed-loop control until the holding pressure stage ends, ensuring that the actual pressure closely tracks the set pressure curve.

[0067] In step S2, the BP neural network compensation model is a three-layer feedforward network. Its input is the original output voltage value of the pressure sensor 12 and the reading of the internal temperature sensor, and its output is the accurate pressure value after nonlinear temperature compensation. This model is obtained by training with a large amount of calibration data covering the pressure range of 0-5000 bar and the temperature range of 30-320°C.

[0068] In step S4, the specific process of adjusting the PID parameters online includes: fuzzifying the precise values ​​of pressure deviation e and deviation change rate ec to obtain the corresponding fuzzy subsets; performing fuzzy inference based on a preset fuzzy rule base to obtain fuzzy subsets of output variables ΔKp, ΔKi, and ΔKd; defuzzifying the output fuzzy subsets to obtain precise PID parameter correction values; and updating the parameters of the current PID controller using the correction values: Kp = Kp_prev + ΔKp, Ki = Ki_prev + ΔKi, Kd = Kd_prev + ΔKd.

[0069] In this embodiment, at the start of the injection phase (time t0), the set pressure P_set rapidly increases from 0 bar to 4000 bar using a ramp function. The system acquires the pressure signal in real time at a frequency of 1 kHz, and obtains P_actual after compensation by a BP neural network.

[0070] In the initial stage of pressure rise (t0-t1), the actual pressure P_actual lags behind P_set, and the pressure deviation e(n) is positive and relatively large. At this time, the adaptive fuzzy PID controller, according to the fuzzy rules, quickly outputs a large proportional gain Kp, causing the PID control quantity u_pid(n) to increase rapidly. Simultaneously, the feedforward compensator detects that the injection speed command is rising sharply and immediately outputs a large positive feedforward control quantity u_ff(n). The superposition of u_pid(n) and u_ff(n) generates a very strong comprehensive control signal u(n), driving the servo valve to open rapidly and wide, causing the injection cylinder to advance rapidly, and the pressure to build up quickly. In contrast, traditional PID controllers, due to their fixed parameters, have a slower response and a significant lag in pressure rise.

[0071] As the pressure approaches the 4000 bar setpoint (t1-t2), the deviation e(n) gradually decreases. Based on the decrease in e(n) and the change in ec(n), the fuzzy controller automatically decreases the value of Kp while simultaneously increasing the derivative gain Kd. The increase in Kd enhances the ability to predict and suppress pressure change trends. This allows the control system to "slow down" as it approaches the target, effectively avoiding pressure overshoot (overshoot) caused by inertia. (Refer to...) Figure 5 The control curve of this invention smoothly approaches the set value.

[0072] Upon entering the holding pressure stage (after t2), the pressure setpoint decreases slightly and remains stable. At this point, pressure fluctuations mainly originate from minor disturbances in the melt compensation flow. When |e| is very small, the fuzzy controller appropriately increases the integral gain Ki to eliminate steady-state residual error. Ultimately, the system stably controls the nozzle pressure near the setpoint, with a fluctuation range far smaller than that of traditional PID control; for example, it can stabilize within a bandwidth of ±20 bar, ensuring the stability of the holding pressure process and the accuracy of product dimensions.

[0073] The implementation principle of this invention is as follows: This invention discloses a nozzle pressure measurement and control system and method capable of repeatedly measuring ultra-high pressure (4000 bar). The system includes: a pressure sensing module 1 using GRAEFF microelectronics technology, equipped with an active cooling and heat insulation structure, enabling stable operation under ultra-high pressure and high temperature environments; a signal processing module integrating a temperature compensation algorithm based on a BP neural network, significantly improving signal accuracy; a core control module 2 combining an adaptive fuzzy PID controller and a feedforward compensator to form an intelligent composite control strategy; and a high-response actuator precisely driving the injection action. The method includes steps such as real-time pressure acquisition and intelligent compensation, pressure deviation calculation, calculation of adaptive fuzzy PID and feedforward composite control quantities, comprehensive output, and pressure regulation. This invention effectively solves key technical problems in ultra-high pressure injection molding, such as large pressure measurement drift, slow control response, and easy oscillation, achieving long-term high-precision measurement and rapid and accurate control under extreme conditions of 4000 bar / 300°C, significantly improving the quality consistency and pass rate of precision injection molded products.

[0074] The embodiments described herein are preferred embodiments of the present invention and are not intended to limit the scope of protection of the present invention. Therefore, all equivalent changes made in accordance with the structure, shape, and principle of the present invention should be covered within the scope of protection of the present invention.

Claims

1. A nozzle pressure measurement and control system capable of repeatedly measuring ultra-high pressure of 4000 bar, characterized in that, include: The pressure sensing module (1) is used to measure the melt pressure at the nozzle of the injection molding machine in real time; The signal processing module is electrically connected to the pressure sensing module (1) and is used to condition and temperature compensate the raw signal output by the pressure sensing module (1). The core control module (2) is communicatively connected to the signal processing module and is used to receive the processed actual pressure value, compare it with the set pressure value, and calculate the control signal based on the deviation. The actuator is connected to the core control module (2) and is used to adjust the injection action of the injection molding machine according to the control signal in order to control the melt pressure at the nozzle.

2. The nozzle pressure measurement and control system capable of repeatedly measuring ultra-high pressure of 4000 bar according to claim 1, characterized in that, The pressure sensing module (1) also includes: An active cooling structure is a double-layer water-cooling jacket surrounding the base of the pressure sensor (12), which is used to introduce constant-temperature cooling water to actively cool the sensor. Passive thermal insulation structure, including high-temperature thermal insulation material filled at the sensor signal leads.

3. The nozzle pressure measurement and control system capable of repeatedly measuring ultra-high pressure of 4000 bar according to claim 1, characterized in that, The signal processing module includes a temperature compensation unit, which uses a neural network-based software algorithm to compensate for the zero-point drift and sensitivity drift of the pressure sensor (12) in real time.

4. The nozzle pressure measurement and control system capable of repeatedly measuring ultra-high pressure of 4000 bar according to claim 1, characterized in that, The core control module (2) includes: An adaptive fuzzy PID controller is used to adjust PID parameters online based on pressure deviation and the rate of change of deviation. A feedforward compensator is used to generate a feedforward control signal based on the injection speed setpoint. The output of the feedforward compensator is superimposed with the output of the adaptive fuzzy PID controller to form a control signal for the actuator.

5. A nozzle pressure measurement and control system capable of repeatedly measuring ultra-high pressure of 4000 bar according to claim 4, characterized in that, In the adaptive fuzzy PID controller, the input variables for fuzzy inference are the pressure deviation e and its rate of change ec, and the output variables are the correction values ​​ΔKp, ΔKi and ΔKd of the PID parameters, where ΔKp is the increment of the proportional coefficient, ΔKi is the increment of the integral coefficient and ΔKd is the increment of the derivative coefficient.

6. A nozzle pressure measurement and control system capable of repeatedly measuring ultra-high pressure of 4000 bar according to claim 1, characterized in that, The actuator is a high-response electro-hydraulic servo valve with a frequency response of not less than 100Hz and its matching servo driver, used to control the injection cylinder of a hydraulic injection molding machine.

7. A nozzle pressure measurement and control system capable of repeatedly measuring ultra-high pressure of 4000 bar according to claim 1, characterized in that, The actuator is a high dynamic response servo motor and precision ball screw drive system, used to control the movement of the screw (34) of the all-electric injection molding machine.

8. A method for controlling nozzle pressure in an ultra-high pressure injection molding machine applied to the system described in any one of claims 1-7, characterized in that, Includes the following steps: S1: System initialization, load the expected pressure curve, and initialize controller parameters; S2: Real-time acquisition of nozzle pressure raw signal and sensor temperature signal, and temperature compensation using a pre-trained BP neural network model to obtain a high-precision actual pressure value P_actual; S3: Calculate the pressure deviation at the current moment e(n) = P_set(n) - P_actual(n) and the rate of change of deviation ec(n) = e(n) - e(n-1); S4: Using an adaptive fuzzy PID controller, the PID parameters are adjusted online based on e(n) and ec(n), and the PID control quantity u_pid(n) is calculated; at the same time, the feedforward compensator calculates the feedforward control quantity u_ff(n) based on the injection speed setpoint. S5: Add the PID control quantity u_pid(n) and the feedforward control quantity u_ff(n) to obtain the comprehensive control signal u(n); S6: Drive the actuator according to the comprehensive control signal u(n) to adjust the injection speed or pressure so that the actual melt pressure at the nozzle tracks the set pressure curve; S7: Repeat steps S2 to S6 during the injection and pressure holding phases until the process is complete.

9. The method for controlling nozzle pressure in an ultra-high pressure injection molding machine according to claim 8, characterized in that, In step S2, the BP neural network compensation model is a three-layer feedforward network. Its input is the original output voltage value of the pressure sensor (12) and the reading of the internal temperature sensor, and its output is the accurate pressure value after nonlinear temperature compensation. The model is obtained by training with a large amount of calibration data covering the pressure range of 0-5000 bar and the temperature range of 30-320°C.

10. The method for controlling nozzle pressure in an ultra-high pressure injection molding machine according to claim 8, characterized in that, In step S4, the specific process of adjusting the PID parameters online includes: The precise values ​​of pressure deviation e and deviation change rate ec are fuzzified to obtain the corresponding fuzzy subsets; Fuzzy inference is performed based on a preset fuzzy rule base to obtain fuzzy subsets of output variables ΔKp, ΔKi, and ΔKd; Defuzzify the output fuzzy subset to obtain accurate PID parameter correction values; Update the parameters of the current PID controller using the correction amount: Kp = Kp_prev + ΔKp, Ki = Ki_prev + ΔKi, Kd = Kd_prev + ΔKd.

Images