Injection control method and injection hydraulic system for a die casting machine

Abstract

The present disclosure relates to a method and system for controlling injection of a die casting machine, and relates to the technical field of die casting machines. The method comprises: acquiring actual displacement, actual pressure and actual velocity of an injection punch in real time; dividing the injection process into multiple process stages based on the actual displacement, and determining preset velocity curves for each stage; acquiring a compensation velocity based on the actual velocity; adaptively planning a target back pressure, combining the target back pressure and the actual pressure, and determining opening degree commands of valves in low and high frequency domains according to the preset velocity curves and the compensation velocity; and according to the current process stage, directing the opening degree commands to each valve. The present disclosure realizes smooth tracking and disturbance-free switching in all working conditions and improves injection stability through segmented planning and double-valve decoupling control.

Description

Technical Field

[0001] This disclosure relates to the field of die casting machine technology, and in particular to an injection control method and injection hydraulic system for a die casting machine. Background Technology

[0002] The injection system is the core working unit of a die-casting machine, possessing speed and pressure control precision within milliseconds, directly determining the internal density and molding quality of the die-cast parts. With the rise of large-scale integrated die-casting processes, mold cavities are becoming increasingly complex, placing extremely high demands on the dynamic response and fluid stability of the injection system. However, injection hydraulic control technology still has physical and control limitations in engineering practice. Summary of the Invention

[0003] According to a first aspect of the present disclosure, a method for injection control of a die-casting machine is proposed. It includes: S1: Real-time acquisition of the actual displacement x(t) of the injection punch collected by the displacement sensor of the die-casting machine and the actual pressure collected by the pressure sensor of the die-casting machine, and determination of the actual velocity v of the injection punch based on the actual displacement x(t). act The actual pressure includes the pressure P of the injection accumulator. s S1: Inlet chamber pressure P1 and outlet chamber pressure P2; S2: Based on the actual displacement x(t), the injection process is divided into multiple continuous process stages, and the preset speed curve v of the injection punch under each process stage is determined. set (t), where t is the injection time; S3: based on the actual velocity v act A combined feedforward and active disturbance rejection control strategy is employed to obtain the high-frequency disturbance rejection compensation velocity Δv; adaptive planning is used to determine the target back pressure P. 2_target Combined with the target back pressure P 2_target And the actual pressure, based on the preset speed curve v set (t) and compensation speed Δv, the valve opening command of at least one valve in the die casting machine in the low frequency domain and high frequency domain is determined by the inverse physics model; S4: According to the current process stage, the valve opening command is directionally distributed to the corresponding valves in the energy storage control cartridge valve, injection control valve group and pressurized oil inlet cartridge valve of the die casting machine, driving the die casting machine to complete the complete injection process. The injection control valve group includes the injection inlet valve group and the injection outlet valve group; the injection inlet valve group includes the injection inlet large diameter cartridge valve and the injection inlet small diameter cartridge valve connected in parallel, and the injection outlet valve group includes the injection outlet large diameter cartridge valve and the injection outlet small diameter cartridge valve connected in parallel.

[0004] The disclosed solution realizes a complete control link from signal acquisition, stage division, composite calculation to valve directional allocation. It improves speed tracking accuracy and anti-interference capability under all operating conditions, laying the foundation for subsequent high-precision injection control.

[0005] In this disclosure, multiple consecutive process stages include: Stage 1 is an energy storage stage, used for storing energy in the injection accumulator and booster accumulator of the die-casting machine, in which v set (t)=0; when time t reaches the required energy storage duration, i.e., t=t1, stage 2 is triggered; stage 2 is a slow stage, in which a cubic polynomial is used to convert the preset speed curve v set (t) smoothly increases from 0 to the wavefront velocity v gather By applying a sealing safety constraint, the injection punch is ensured to reach the set sealing safety position x at the soup inlet. seal That is, t=t s At that time, the actual speed v act Not greater than the set sealing safety speed v seal When t reaches t1 + T acc1 Phase 3 is triggered by time, where t1 is the start time of the slow phase; T acc1 This represents the total duration of the slow phase; Phase 3 is a fast phase, where the preset speed curve v... set (t) arrival wavefront velocity v gather Then, with v gather Constantly push out molten metal for venting; when the actual displacement x(t) reaches the ingate position x gate That is, stage 4 is triggered at t=t2; stage 4 is the acceleration stage, in which a globally continuous fifth-degree polynomial is used to accelerate the preset velocity curve v. set (t) from v gather Rapidly pull up to high compression velocity v fast This causes the injection punch to rapidly fill the cavity with molten metal; when t reaches t2 + T acc2 After triggering phase 5, t2 is the start time of the acceleration phase, T acc2 Phase 5 is the total duration of the acceleration phase; Phase 5 is the rapid phase, with a preset speed curve v. set (t) Maintain high injection speed v fast Constant; relying on the active disturbance rejection expansion state observer, the total disturbance estimate and the differential rate of change of the total disturbance estimate of the die casting machine during the die casting process are estimated in real time. When the injection punch breaks through the set deceleration safety position x safe And the differential rate of change exceeds the precursor threshold ε pre That is, stage 6 is triggered at t=t3; stage 6 is the deceleration stage, in which a fifth-order polynomial with a lower speed limit bias is used to adjust the preset speed curve v. set (t) Implement tolerance braking, and set the preset speed curve v set (t) from fast injection velocity v fast Reduce to a safe collision speed v safe v safe >0; when t reaches t3+T dec Afterwards, maintain a safe collision speed vsafe Phase 7 is triggered when t4 is reached, and t3 is the start time of the deceleration phase. dec Phase 7 is the total duration of the deceleration phase; Phase 8 is the boost phase, with a preset speed curve v. set (t) Forced to zero.

[0006] The disclosed solution fundamentally avoids the process defects of premature undercasting or late flashing in traditional equipment, and improves the molding consistency and yield of integrated die-cast parts.

[0007] In this disclosure, the preset velocity curve v set The key spatial nodes and velocity thresholds in (t) include the ingate location x. gate Wavefront velocity v gather High-speed injection velocity v fast At least one of the following: Ingate location x gate The position of the injection punch when it propels the molten metal to the ingate in a rapid phase is represented as:

[0008] x gate =(V biscuit +V runner ) / A chamber (1)

[0009] Among them, A chamber V is the cross-sectional area of ​​the injection chamber containing the molten metal; biscuit V represents the volume of the casting spool; runner For the volume of the horizontal gating system; wavefront velocity v gather The velocity set to prevent air entrapment in the molten metal during the rapid propulsion phase of the injection plunger is expressed, based on the shallow water wave critical velocity theory, as follows:

[0010] (2)

[0011] Where f is the injection chamber filling rate; D is the injection chamber diameter; g is the gravitational acceleration; k is the wavefront coefficient; and v is the rapid injection velocity. fast The speed set to ensure the molten metal fills the mold cavity properly during the two rapid phases of the injection plunger's propulsion is expressed as follows, based on the filling energy equivalent equation:

[0012] v fast =[V total / (A gate ×t fill )]×(A gate / A chamber (3)

[0013] Among them, V total A represents the total volume of the casting; gatet is the total cross-sectional area of ​​the ingate. fill Time required to fill the target area.

[0014] The disclosed solution enables the preset speed curve to automatically adapt to the volume, cross-sectional area, and filling time requirements of different molds, achieving intelligent planning of one curve per mold. This reduces the process debugging cycle and parameter adjustment difficulty, while theoretically avoiding typical injection defects such as air entrapment, insufficient filling, and flash.

[0015] In this disclosure, for stage 2, a preset velocity curve v is determined. set (t) includes: selecting a cubic polynomial:

[0016] v(τ1) = C0 + C1τ1 + C2τ1 2 + C3τ1 3 (4)

[0017] Where C0, C1, C2, and C3 are the coefficients of a cubic polynomial, and τ1 is the normalized time independent variable for stage 2, τ1=(t-t1) / T acc1 v(τ1) is the velocity dependent variable corresponding to stage 2;

[0018] Determine the physical boundary conditions for the slow-pressure injection phase, including: starting from rest: when τ1=0, v(0)=0; reaching the wavefront velocity: when τ1=1, v(1)= v gather ; To avoid impact and splashing at the start, the acceleration is 0: when τ1=0, a(0)=0; At the end, smoothly transition to a uniform pushing state without waves, the acceleration is 0: when τ1=1, a(1)=0; Substitute the physical boundary conditions of the slow-pressure injection section into the derivative of the cubic polynomial to obtain the general coefficients of the cubic polynomial, and obtain the preset velocity curve v of stage 2. set (t):

[0019] v set (t)= v gather ×(3τ1 2 -2τ1 3 (5)

[0020] Among them, T acc1 It is determined based on the actual displacement x(t) and the sealing safety constraint. The actual displacement x(t) is determined according to the velocity integral formula, expressed as:

[0021] (6)

[0022] The sealing safety constraint is: when there is a time point t s Make the actual displacement x(t) s )=x seal At that time, the current speed v must be strictly satisfied. set (ts )≤v seal .

[0023] The disclosed solution eliminates fluid splashing and impact surges at the moment of initiation, while avoiding hydraulic shocks caused by sudden acceleration changes, providing stable and safe initial flow field conditions for subsequent high-speed injection, and significantly improving the operational stability of the low-speed injection section.

[0024] In this disclosure, for stage 4, a preset velocity curve v is determined. set (t) includes: selecting a fifth-degree polynomial:

[0025] v(τ2)=C0+C1τ2+C2τ2 2 + C3τ2 3 + C4τ2 4 + C5τ2 5 (7)

[0026] Where C0, C1, C2, C3, C4, and C5 are the coefficients of the fifth-order polynomial, and τ2 is the normalized time independent variable for stage 4, τ2=(t-t2) / T acc2 v(τ2) is the velocity dependent variable corresponding to stage 4;

[0027] Determine the physical boundary conditions for the acceleration phase, including: the starting point of the acceleration phase connects to the wavefront velocity of the fast phase: when τ2=0, v(0)= v gather When the target rapid injection velocity is reached at the endpoint: τ2=1, v(1)= v fast ; No impact at start-up, acceleration is 0: when τ2=0, a(0)=0; No overshoot oscillation before entering the fast injection stage, acceleration is 0: when τ2=1, a(1)=0; Smooth torque at the moment the valve core opens, jerk is 0: when τ2=0, j(0)=0; Smooth torque at the moment the valve core stops, jerk is 0: when τ2=1, j(1)=0;

[0028] Substituting the physical boundary conditions of the acceleration section into the derivative of the fifth-order polynomial, we can obtain the coefficients of the general formula of the fifth-order polynomial, and thus obtain the preset velocity curve v of stage 4. set (t):

[0029] v set (t) = v gather +( v fast - v gather ) ×(10τ2 3 -15τ2 4 +6τ2 5 (8)

[0030] Among them, T acc2It is determined using a dynamic adaptive criterion based on the target valve opening ratio, expressed as:

[0031] (9)

[0032] In the formula, T valve_step v is the full-scale step response time of the system's servo cartridge valve; max k is the maximum design speed of the injection system. acc A safety tracking factor is set between 1.2 and 1.5; T min The minimum acceleration time limit to prevent high-frequency excitation.

[0033] The disclosed scheme employs a fifth-order polynomial satisfying six boundary conditions (position, velocity, acceleration, and jerk) during the acceleration phase, achieving a globally continuous and smooth increase from the wavefront velocity to the fast injection velocity. Simultaneously, a dynamic acceleration duration adaptive criterion based on the valve core response limit is introduced to ensure that the target velocity change rate does not exceed the physical tracking capability of the servo valve. This achieves the goal of extreme acceleration without loss of control, significantly improving the response speed and control accuracy of the fast injection phase.

[0034] In this disclosure, when the injection plunger exceeds the set deceleration safety position x safe And the differential rate of change exceeds the precursor threshold ε pre That is, at t=t3, stage 6 is triggered, which includes: during the operation of stage 5, reading the total disturbance estimate z2 estimated by the autonomous disturbance rejection extended state observer in real time; physically mapping the total disturbance estimate z2 to the comprehensive operating resistance of the injection process; performing a first-order low-pass filter on the total disturbance estimate z2 and then performing differential approximation to extract the comprehensive disturbance drastic rate. When the actual displacement x(t) > x safe And detected >ε pre When the molten metal has filled the main cavity of the mold, it is determined that the molten metal has filled the main cavity of the mold, triggering the entry into stage 6.

[0035] The disclosed solution makes deceleration braking no longer dependent on a fixed position or time, but truly responds to the actual flow resistance changes during the filling process, significantly improving the robustness and adaptability of deceleration triggering, and avoiding the over-braking or under-braking problems caused by flow resistance fluctuations in traditional injection.

[0036] In this disclosure, for stage 6, a preset velocity curve v is determined. set (t) includes: selecting a fifth-degree polynomial:

[0037] v(τ3)=C0+C1τ3+C2τ3 2 + C3τ3 3 + C4τ3 4 + C5τ3 5 (10)

[0038] Where C0, C1, C2, C3, C4, and C5 are the coefficients of the fifth-order polynomial, and τ3 is the normalized time independent variable for stage 6, τ3=(t-t3) / T dec v(τ3) is the velocity dependent variable corresponding to stage 4;

[0039] Determine the physical boundary conditions of the deceleration section, including: the initial seam connection speed of the fast injection: when τ3=0, v(0)= v fast When the safe collision speed is reached at the endpoint: τ3=1, v(1)= v safe ; The starting point transitions smoothly without sudden water hammer impact. When the acceleration is 0: τ3=0, a(0)=0; The ending point smoothly completes the braking action. When the acceleration is 0: τ3=1, a(1)=0; The torque is smooth at the moment the valve core begins to close. When the jerk is 0: τ3=0, j(0)=0; The torque is smooth at the moment the valve core closes. When the jerk is 0: τ3=1, j(1)=0;

[0040] Substituting the physical boundary conditions of the deceleration section into a fifth-order polynomial yields the general coefficients of the fifth-order polynomial, thus obtaining the preset velocity curve v for stage 6. set (t):

[0041] v set (t) = v safe +( v fast - v safe )×(1-(10τ3 3 -15τ3 4 +6τ3 5 )) (11)

[0042] Among them, T dec Rigidity calculation is performed using a braking integral matching criterion based on mold geometric tolerance: the total volume of the mold overflow channel and venting channel is obtained, and the equivalent safe braking displacement S is calculated in reverse. rem According to the preset velocity curve v of stage 6 set The integral of (t) is obtained from the fact that the displacement is always equal to 0.5(v). fast + v safe )×T dec Based on the characteristics, we obtain:

[0043] T dec =2×S rem / ( v fast + v safe (12)

[0044] After deceleration in stage 6, the preset speed curve v set (t) Maintain a safe collision speed v safe Up to stage 7.

[0045] The disclosed solution strictly binds the deceleration process with the actual tolerance space of the mold, which avoids flash and mold damage caused by braking too late, and also prevents under-casting caused by braking too early. At the same time, it provides a stable time window for the pre-filling of the booster cylinder.

[0046] In this disclosure, S3 includes: in the feedforward solution branch, the preset velocity curve v set (t) is directly input into the inverse physics model, and the target back pressure P is... 2_target Substituting the actual pressure into the inverse physics model according to different process stages and valve differences, we calculate the low-frequency basic valve opening command u that satisfies macroscopic trajectory tracking and includes the expected macroscopic damping back pressure. base Low-frequency basic valve opening command u base Including the low-frequency basic valve opening command u on the oil inlet side in_base Oil outlet side low-frequency basic valve opening command u out_base In the active disturbance rejection solution branch, the preset velocity curve v is used. set (t) and actual velocity v act A comparison is performed to generate a velocity error, which is then input into the active disturbance rejection regulator in the virtual command domain. The total disturbance estimate is then calculated in real time using an extended state observer, and combined with a linear error feedback control law, the high-frequency disturbance rejection compensation velocity Δv is calculated. Δv is then input into the inverse physics model, and the target back pressure P is calculated. 2_target Substituting the actual pressure into the inverse physics model according to different process stages and different valves, the high-frequency compensated valve opening command u is calculated. comp High-frequency compensated valve opening command u comp Including the high-frequency compensation valve opening command u on the oil inlet side in_comp Oil outlet side high-frequency compensation valve opening command u out_comp Among them, the target back pressure P 2_target The actual pressure is substituted into the inverse physics model according to different process stages and different valves, including: calculating the low-frequency basic valve opening command u on the oil side in the slow stage and the fast stage. out_base Oil outlet side high-frequency compensation valve opening command u out_comp At that time, the target back pressure P 2_target Substituting into the inverse physics model; during the acceleration and second rapid phases, the low-frequency basic valve opening command u on the oil side is calculated. out_base At that time, the target back pressure P 2_target Substitute the actual pressure into the inverse physics model; for the rest, substitute the actual pressure into the inverse physics model.

[0047] This disclosed solution constructs a dual-path parallel composite control architecture consisting of a feedforward calculation branch and an active disturbance rejection (ADRC) calculation branch. By combining this with differentiated substitution rules for target back pressure and actual pressure, it achieves physical separation between low-frequency trajectory tracking and high-frequency disturbance suppression. The feedforward branch is responsible for macroscopic velocity tracking and back pressure establishment, while the ADRC branch is responsible for real-time cancellation of unmodeled friction and nonlinear filling resistance. This resolves the logical conflict in traditional control where velocity tracking and pressure establishment occur on the same valve core, significantly improving velocity tracking accuracy and disturbance rejection capability across all operating conditions.

[0048] In this disclosure, the high-frequency disturbance rejection compensation velocity Δv is calculated, including: based on the actual velocity v using an extended state observer. act Based on the high-frequency disturbance rejection compensation velocity Δv0 of the previous control cycle, the real-time discrete update equation is determined as follows:

[0049] e obs =z1-v act (13)

[0050] z1 = z1 + h × (z2 - β1 × e) obs +b0×△v0) (14)

[0051] z2 = z2 + h × (-β2 × e) obs (15)

[0052] In the formula, z1 is the observed and tracked value of the actual velocity; e obs β1 is the observation error; h is the discrete sampling step size; β1 and β2 are the observer gain parameters; b0 is the system virtual control gain; the total disturbance estimate z2 reflects the mechanical friction of the injection punch and the nonlinear filling flow resistance of the molten metal in the mold cavity;

[0053] Based on the preset velocity curve v set The tracking error e is calculated using (t) and the observed tracking value z1. v = v set (t)-z1;

[0054] Based on the tracking error, a linear error feedback control law is used to generate a virtual primary control quantity u0=k. p ×e v , where k p For proportional gain;

[0055] Based on the total disturbance estimate z2, feedforward compensation is performed on the virtual primary control variable to calculate the high-frequency disturbance rejection compensation speed Δv:

[0056] △v = (u0-z2) / b0 (16)

[0057] This disclosed scheme achieves real-time, high-precision estimation of the mechanical friction of the injection punch and the flow resistance of the molten metal filling by placing the Extended State Observer (LESO) in the virtual command domain and using the actual speed and the compensated speed of the previous cycle as observation inputs. Combined with linear error feedback and feedforward compensation, this method can generate the compensated speed before system disturbances occur, significantly improving the control's anticipation and robustness, significantly suppressing high-frequency speed glitches in the fast injection phase, and achieving micron-level high-precision speed tracking.

[0058] In this disclosure, the low-frequency basic valve opening command u base and high-frequency compensated valve opening command u comp The inverse physics model is obtained through the following steps:

[0059] Obtaining the inverse physics model:

[0060] (17)

[0061] In the formula, u represents the valve opening command, including the low-frequency basic valve opening command u on the oil inlet side. in_base Oil outlet side low-frequency basic valve opening command u out_base , Inlet side high-frequency compensation valve opening command u in_comp Oil outlet side high-frequency compensation valve opening command u out_comp A represents the corresponding hydraulic cylinder working area; V cmd Substitute the target speed; C d K is the flow coefficient; v ρ is the valve's rated flow area gradient; ρ is the hydraulic oil density; ΔP is the driving pressure difference.

[0062] Determine the variable substitution rule: For the effective area A=A1 on the injection inlet side, the driving pressure difference ΔP=P s -P1 holds true at all times, where P s P1 is the pressure of the injection accumulator, and P2 is the pressure of the inlet chamber. For the effective area of ​​the injection outlet side, A=A2, the driving pressure difference ΔP=P 2_ref –P t , where P t Let P be the oil return pressure from the oil tank. t =0; Target back pressure P on the oil outlet side 2_target The oil outlet pressure P2 is substituted into the inverse physics model according to different usage scenarios; the low-frequency basic valve opening command u on the oil inlet side is calculated. in_base Oil outlet side low-frequency basic valve opening command u out_base At that time, V cmd Substitute v set (t), calculate the opening command u of the high-frequency compensation valve on the oil inlet side. in_comp Oil outlet side high-frequency compensation valve opening command u out_compAt that time, V cmd Substitute Δv; according to the inverse physics model and variable substitution rules, the target back pressure P is... 2_target The directional input is fed into the inverse physics model to replace the oil outlet pressure P2, and the low-frequency basic valve opening command u is calculated. base and high-frequency compensated valve opening command u comp .

[0063] This disclosed solution establishes a unified mathematical model for throttling inverse kinematics and clarifies the variable rules for substituting the constant pressure difference on the inlet side, the target back pressure on the outlet side, and the actual pressure differences. This achieves high-precision, low-delay mapping of speed commands to valve opening. This method enables both feedforward control and ADRC compensation to be executed efficiently under the same physical model. Simultaneously, it provides personalized pressure substitution strategies for different stages and control branches, achieving complete decoupling of speed and pressure, and significantly improving the system's control consistency and engineering feasibility.

[0064] In this disclosure, the criteria for generating the target back pressure include: in stage 3, the minimum damping back pressure lower limit is derived based on the sudden change in static and dynamic frictional force of the system, and the maximum allowable back pressure upper limit is derived by back-calculating based on the linear dead zone safety opening constraint of the oil outlet valve port; within the effective physical boundary formed by the minimum damping back pressure lower limit and the maximum allowable back pressure upper limit, a static constant high pressure is selected as the target back pressure for stage 3; in stage 5, the target back pressure lower limit is derived based on the critical vibration damping criterion for suppressing velocity vibration; a constant low pressure equal to or higher than the target back pressure lower limit is selected as the target back pressure for stage 5; in stages 2 and 4, step switching of the target back pressure is prohibited, and the pre- and pre- The speed curve is forcibly bound to a smooth mapping: In stage 2, the preset speed curve of stage 2 is reused to smoothly establish the target back pressure of stage 2; In stage 4, the preset speed curve of stage 4 is reused, and an absolutely smooth dynamic unloading mapping is performed from static constant high pressure to constant low pressure to determine the target back pressure of stage 4; In stage 6, the active planning constraint of the target back pressure is released, the target back pressure is switched to the oil outlet chamber pressure P2 and substituted into the inverse physics model, and the oil outlet side valve is controlled to strictly reduce the oil discharge opening synchronously according to the descending deceleration preset speed command. The inertial kinetic energy of the system is absorbed by the valve port throttling damping, so that the oil outlet chamber passively establishes a high-level braking back pressure as the target back pressure.

[0065] The disclosed scheme achieves dynamic adaptive control of back pressure. High back pressure is used to eliminate creep during the slow phase, low back pressure is used to reduce energy loss during the fast phase, and the actual pressure is actively switched during the deceleration phase to achieve flexible braking.

[0066] In this disclosure, S4 includes: in stages 2 and 3, the valve opening commands of the forced injection inlet large-diameter cartridge valve and the forced injection outlet large-diameter cartridge valve are zero; the total command, which is the superposition of the inlet-side low-frequency basic valve opening command and the inlet-side high-frequency compensated valve opening command, is allocated to the forced injection inlet small-diameter cartridge valve; simultaneously, the outlet-side low-frequency basic valve opening command u out_base With the high-frequency compensation valve opening command u on the oil outlet side out_comp The superimposed total command is assigned to the injection outlet small-diameter cartridge valve; in stages 4 and 5, the low-frequency basic valve opening command on the inlet side is directionally assigned to the injection inlet large-diameter cartridge valve, and the low-frequency basic valve opening command on the outlet side is assigned to the injection outlet large-diameter cartridge valve. out_base Directional allocation to the large-diameter cartridge valve for injection outlet; simultaneously, directional allocation of the high-frequency compensating valve opening command on the inlet side to the small-diameter cartridge valve for injection outlet, and directing the high-frequency compensating valve opening command on the outlet side... out_comp Directional allocation to the injection outlet small-diameter cartridge valve; in stage 6, maintain the enabled state of the injection inlet large-diameter cartridge valve and the injection outlet large-diameter cartridge valve and continue to receive the inlet side low-frequency basic valve opening command u. in_base Oil outlet side low-frequency basic valve opening command u out_base Strictly following the fifth-order polynomial trajectory of the deceleration phase, smooth flow contraction tolerance braking is implemented until the kinetic energy decays to the set low-speed tailing threshold. At this point, the valve opening commands of the injection inlet large-diameter cartridge valve and the injection outlet large-diameter cartridge valve are forcibly cut off. During this period, the injection inlet small-diameter cartridge valve and the injection outlet small-diameter cartridge valve continue to receive high-frequency compensation valve opening commands from the inlet side and the outlet side to continuously perform high-frequency compensation. Simultaneously, output is sent to the booster inlet cartridge valve. A small opening signal is used to pre-charge the booster cylinder with fluid; in stage 7, the valve opening commands of the large-diameter injection inlet cartridge valve, the large-diameter injection outlet cartridge valve, and the small-diameter injection inlet cartridge valve are forced to be zero; a fully open command is output to the booster inlet cartridge valve to trigger the zero-hysteresis transient establishment of the booster pressure; at the same time, the speed tracking control of the small-diameter injection outlet cartridge valve is forcibly released, causing the small-diameter injection outlet cartridge valve to switch to the micro-motion compensation and clearance mode, and a fully open command is output to the small-diameter injection outlet cartridge valve.

[0067] This disclosed solution uses a state machine to dynamically reorganize and directionally allocate valve opening commands at different process stages, achieving full-process hardware scheduling: independent control of small valves in the slow stage, flow supply from large valves and disturbance compensation from small valves in the fast stage, pre-filling in the deceleration stage, and yielding and shrinkage compensation in the pressurization stage. This allocation logic enables each valve to perform its function and work collaboratively throughout the entire process, achieving zero-hysteresis pressure build-up and high-precision shrinkage compensation at the hardware execution level, significantly improving the internal density and molding quality of large die-cast parts.

[0068] According to a second aspect of the present disclosure, an injection hydraulic system for a die-casting machine is provided. It includes: a controller for executing the method of any of the first aspects of the present disclosure; a power and energy storage unit including an oil tank, a hydraulic source, an injection accumulator, a booster accumulator, and an energy storage control cartridge valve; the oil outlet of the hydraulic source is connected to the injection accumulator and the booster accumulator respectively through the energy storage control cartridge valve; an execution unit including an injection cylinder, an injection punch and a check valve disposed within the injection cylinder, and a booster cylinder; and an injection control valve group including an injection inlet valve group and an injection outlet valve group; the injection inlet valve group includes a large-diameter injection inlet cartridge valve and a small-diameter injection inlet cartridge valve connected in parallel, and connected to the injection accumulator. The system includes: an injection accumulator and an injection cylinder; an injection outlet valve assembly consisting of a large-diameter injection outlet cartridge valve and a small-diameter injection outlet cartridge valve connected in parallel, and connected between the injection cylinder outlet and the oil tank; a booster control valve assembly consisting of a booster inlet cartridge valve connected between the booster accumulator and the booster cylinder outlet, with the booster cylinder outlet directly connected to the oil tank; and a status detection unit consisting of an accumulator pressure sensor located at the injection accumulator, an inlet chamber pressure sensor located at the injection cylinder outlet, an outlet chamber pressure sensor located at the outlet, and a displacement sensor connected to the injection punch.

[0069] This disclosed system constructs a complete hydraulic system including a power and energy storage unit, an execution unit, an injection control valve group, a pressure boosting control valve group, and a status detection unit. It employs a valve group structure with parallel large and small diameter servo valves for injection inlet and outlet, achieving independent flow control on the inlet and outlet sides of the injection cylinder. This architecture provides hardware support for the injection control method of die-casting machines, resolving the physical contradiction between flow rate and frequency response in traditional single-valve systems, and significantly improving the system's control freedom, response bandwidth, and engineering reliability. Attached Figure Description

[0070] To more clearly illustrate the technical solutions in the embodiments of this disclosure, the accompanying drawings required for the description of the embodiments are introduced below. The following drawings are only some embodiments of this disclosure and do not impose specific limitations on the protection scope of this disclosure.

[0071] Figure 1 This is a schematic flowchart of an injection control method for a die-casting machine according to an embodiment of the present disclosure;

[0072] Figure 2 This is a schematic diagram of the architecture of an injection hydraulic system for a die-casting machine according to an embodiment of the present disclosure;

[0073] Figure 3 This is a schematic diagram of a hydraulic injection system for a die-casting machine according to an embodiment of the present disclosure;

[0074] Figure 4This is a control flowchart of a die-casting machine injection system provided according to an embodiment of the present disclosure;

[0075] Figure 5 This is a schematic diagram of a design principle for a target injection speed curve of a die-casting machine, provided according to an embodiment of this disclosure.

[0076] Figure 6 This is a schematic diagram of a high-precision tracking control principle for the injection speed of a die-casting machine, provided according to an embodiment of this disclosure.

[0077] Figure 7 This is a schematic diagram of valve command decoupling and directional allocation in a die-casting machine injection system according to an embodiment of this disclosure.

[0078] Figure label:

[0079] 1. Large-diameter servo valve for injection oil outlet; 2. Small-diameter servo valve for injection oil outlet; 3. Large-diameter servo valve for injection oil inlet; 4. Small-diameter servo valve for injection oil inlet; 5. Cartridge valve for booster accumulator energy storage control; 6. Cartridge valve for injection accumulator energy storage control; 7. Cartridge valve for booster oil inlet; 8. Booster accumulator; 9. Injection accumulator; 10. Injection cylinder; 11. Injection punch; 12. Check valve; 13. Booster cylinder; 14. Displacement sensor; 15. Oil outlet chamber pressure sensor; 16. Oil inlet chamber pressure sensor; 17. Accumulator pressure sensor; 18. Hydraulic source; 19. Oil tank; 20. Controller. Detailed Implementation

[0080] This disclosure presents an injection control method and injection hydraulic system for a die-casting machine to solve technical problems such as trajectory solidification, strong coupling interference between pressure and speed, and pressure build-up hysteresis in existing injection systems.

[0081] This disclosure is not exhaustive, but merely illustrative of some embodiments, and is not intended to limit the scope of protection of this disclosure. Unless otherwise specified, each step, component, part, device, or apparatus in a particular embodiment can be implemented as an independent embodiment, and these steps, components, parts, devices, and apparatuses can be arbitrarily combined. For example, the solution after removing some steps, components, parts, devices, or apparatuses in a particular embodiment can also be implemented as an independent embodiment, and the order of steps in a particular embodiment can be arbitrarily interchanged. Furthermore, the optional implementation methods in a particular embodiment can be arbitrarily combined; moreover, the embodiments can be arbitrarily combined. For example, some or all steps, components, parts, devices, and apparatuses of different embodiments can be arbitrarily combined, and a particular embodiment can be arbitrarily combined with the optional implementation methods of other embodiments. In all embodiments of this disclosure, unless otherwise specified or logically conflicting, the terminology and / or descriptions between the embodiments are consistent and can be mutually referenced. Technical features in different embodiments can be combined to form new embodiments based on their inherent logical relationships.

[0082] The terminology used in the embodiments of this disclosure is for the purpose of describing particular embodiments only and is not intended to limit the scope of this disclosure.

[0083] Among related technologies, pressure injection hydraulic control technology has many physical and technical defects, mainly reflected in the following aspects: the trajectory planning is fixed, making it impossible to achieve process self-adaptation; the parameter setting is extremely dependent on human experience, resulting in blind deceleration and braking timing; the pressure and speed are strongly coupled, and the control process has "internal friction and mutual struggle", causing the system to fall into serious internal friction and mutual struggle and high-frequency oscillation, resulting in complete inaccuracy of flow control; the hardware execution is rough, and the action is mismatched under extreme working conditions.

[0084] Therefore, in order to solve the above problems, this solution proposes an injection control method and injection hydraulic system for a die casting machine. Based on a dual-valve independent control architecture, it adopts adaptive stage division, feedforward-ADRC composite control, dynamic back pressure planning, and frequency domain decoupling and directional allocation technology for large and small valves. This achieves smooth tracking without disturbance under all working conditions, zero-hysteresis pressure building, and high-precision micro-motion feeding, significantly improving the forming quality and process stability of large die castings.

[0085] The injection control method and injection hydraulic system of the die-casting machine provided in this disclosure will be described in detail below with reference to the accompanying drawings.

[0086] The injection control method proposed in this disclosure can be executed by the die-casting machine / die-casting hydraulic system, or by the controller inside the die-casting machine, or by a controller that has a control relationship with the die-casting machine, and this disclosure does not limit it.

[0087] Figure 1 This is a schematic flowchart of an injection control method for a die-casting machine according to an embodiment of this disclosure, as shown below. Figure 1 As shown, the method includes the following steps.

[0088] Step S1: Real-time acquisition of the actual displacement x(t) of the injection punch collected by the displacement sensor of the die-casting machine and the actual pressure collected by the pressure sensor of the die-casting machine, and determination of the actual velocity v of the injection punch based on the actual displacement x(t). act The actual pressure includes the pressure P of the injection accumulator. s The inlet pressure P1 and the outlet pressure P2.

[0089] The displacement sensor of the die-casting machine can be installed on the die-casting machine and connected to the injection punch to collect the actual displacement signal of the injection punch in real time.

[0090] In some embodiments, the displacement sensor may be a linear displacement sensor such as a magnetic grating ruler or an optical grating ruler, which is installed on the injection cylinder or injection component, and its movable end is linked with the injection punch to provide feedback on the real-time position of the injection punch during the injection process.

[0091] The pressure sensors may include accumulator pressure sensors, inlet chamber pressure sensors, outlet chamber pressure sensors, etc. The accumulator pressure sensor is installed at the injection accumulator to monitor the pressure of the injection accumulator in real time, i.e., to collect the injection accumulator pressure P. s The inlet pressure sensor is installed at the inlet end of the injection cylinder to monitor the pressure of the inlet chamber in real time, thus obtaining the inlet pressure P1; the outlet pressure sensor is installed at the outlet end of the injection cylinder to monitor the pressure of the outlet chamber in real time, thus obtaining the outlet pressure P2.

[0092] In some embodiments, a displacement sensor can be used to collect the actual displacement x(t) of the injection punch in real time. This actual displacement can be a displacement amount that reflects the change of the injection punch over time. The actual velocity v of the injection punch can be obtained by differentiating or calculating the actual displacement x(t). act .

[0093] For example, the controller of the die-casting machine collects the actual displacement signal of the injection punch from the displacement sensor in real time, and then calculates the actual speed signal of the injection punch through differentiation.

[0094] Step S2, based on the actual displacement x(t), divide the injection process into multiple consecutive process stages, and determine the preset speed curve v of the injection punch in each process stage. set (t), where t is the injection time.

[0095] Based on the actual displacement, the process stage of the injection process can be determined according to the specific position of the injection punch, and the speed at which the injection punch should move in each stage can be calculated, that is, the preset speed curve of each process stage can be determined.

[0096] Each of the multiple process stages has a corresponding preset speed curve, and the preset speed curves of each stage can be smoothly connected.

[0097] In some embodiments, multiple consecutive process stages include: Stage 1 is an energy storage stage, used to wait for the injection accumulator and booster accumulator of the die-casting machine to store energy, in which v set (t)=0; when time t reaches the required energy storage duration, i.e., t=t1, stage 2 is triggered; stage 2 is a slow stage, in which a cubic polynomial is used to convert the preset speed curve v set (t) smoothly increases from 0 to the wavefront velocity v gatherBy applying a sealing safety constraint, the injection punch is ensured to reach the set sealing safety position x at the soup inlet. seal That is, t=t s At that time, the actual speed v act Not greater than the set sealing safety speed v seal When t reaches t1+T acc1 Phase 3 is triggered by time, where t1 is the start time of the slow phase; T acc1 This represents the total duration of the slow phase; Phase 3 is a fast phase, where the preset speed curve v... set (t) arrival wavefront velocity v gather Then, with v gather Constantly push out molten metal for venting; when the actual displacement x(t) reaches the ingate position x gate That is, stage 4 is triggered at t=t2; stage 4 is the acceleration stage, in which a globally continuous fifth-degree polynomial is used to accelerate the preset velocity curve v. set (t) from v gather Rapidly pull up to high compression velocity v fast This causes the injection punch to rapidly fill the cavity with molten metal; when t reaches t2 + T acc2 After triggering phase 5, t2 is the start time of the acceleration phase, T acc2 Phase 5 is the total duration of the acceleration phase; Phase 5 is the rapid phase, with a preset speed curve v. set (t) Maintain high injection speed v fast Constant; relying on the active disturbance rejection expansion state observer, the total disturbance estimate and the differential rate of change of the total disturbance estimate of the die casting machine during the die casting process are estimated in real time. When the injection punch breaks through the set deceleration safety position x safe And the differential rate of change exceeds the precursor threshold ε pre That is, stage 6 is triggered at t=t3; stage 6 is the deceleration stage, in which a fifth-order polynomial with a lower speed limit bias is used to adjust the preset speed curve v. set (t) Implement tolerance braking, and set the preset speed curve v set (t) from fast injection velocity v fast Reduce to a safe collision speed v safe v safe >0; when t reaches t3+T dec Afterwards, maintain a safe collision speed v safe Phase 7 is triggered when t4 is reached, and t3 is the start time of the deceleration phase. dec Phase 7 is the total duration of the deceleration phase; Phase 8 is the boost phase, with a preset speed curve v. set (t) Forced to zero.

[0098] Among them, multiple consecutive process stages can determine whether to enter the corresponding process stage based on time and actual displacement. When the entry conditions set for the corresponding stage are met, the stage is entered; otherwise, the stage is not entered, and the entry conditions for the next stage are judged until the entry conditions for all stages are not met, at which point the current control process is exited.

[0099] In some embodiments, stage 1 is a preparation stage for waiting for the injection accumulator and booster accumulator to store energy. Specifically, the hydraulic source pressurizes the injection accumulator and booster accumulator through the energy storage control cartridge valve until a set pressure value is reached. During this stage, the injection punch remains stationary, and the preset speed command v... set (t)=0.

[0100] In this embodiment, the required duration t1 for energy storage can be set for stage 1. When the duration t of energy storage is less than or equal to t1, the entry condition for stage 1 is met, and the energy storage stage is entered. The energy storage control cartridge valve is opened until the duration t reaches the required duration of energy storage, i.e., t=t1, at which point stage 2 is triggered. That is, the judgment condition for entering stage 2 is when t exceeds t1.

[0101] In some embodiments, stage 2 is the stage in which the injection punch smoothly starts from a stationary state and slowly propels the molten metal forward to the sealing position of the inlet, i.e., the slow stage. The entry condition for stage 2 can be t1 < t.

[0102] In this embodiment, the total duration T of the slow phase can be set for phase 2. acc1 That is, to make a judgment based on time t, when t≤t1+T acc1 This indicates that the current time is in stage 2, and the following needs to be executed: use a cubic polynomial to convert the preset velocity curve v set (t) smoothly increases from 0 to the wavefront velocity v gather By applying a sealing safety constraint, the injection punch is ensured to reach the set sealing safety position x at the soup inlet. seal That is, t=t s At that time, the actual speed v act Not greater than the set sealing safety speed v seal .

[0103] In this embodiment, stage 2 can be achieved by smoothly increasing the speed from 0 to the wavefront speed, allowing the molten metal to advance steadily and preventing fluid splashing at the start. Simultaneously, a sealing safety constraint is applied to ensure that the actual speed of the punch when it reaches the sealing safety position at the broth outlet does not exceed the set sealing safety speed. Stage 3 is triggered after the slow speed period ends.

[0104] Wherein, wavefront velocity v gatherThe velocity set to prevent air entrapment in the molten metal during the rapid propulsion phase of the injection plunger is expressed, based on the shallow water wave critical velocity theory, as follows:

[0105] (2)

[0106] Where f is the injection chamber filling rate; D is the inner diameter of the injection chamber; g is the gravitational acceleration; and k is the wavefront coefficient.

[0107] In this embodiment, the injection chamber filling rate can be the proportion of the volume of liquid metal in the injection chamber to the total volume of the injection chamber, reflecting the liquid level height of the liquid metal in the injection chamber and determining the wavefront flow state; the inner diameter of the injection chamber determines the cross-sectional area of ​​the injection chamber and affects the flow behavior of the liquid metal; gravitational acceleration is used to describe the effect of gravity on the free surface of the liquid metal; the wavefront coefficient is used as a correction factor to adapt to different alloy materials, temperatures and process conditions.

[0108] The filling rate of the injection chamber can be the quotient of the volume of molten metal and the total volume of the injection chamber. The higher the filling rate, the higher the liquid level of the molten metal in the injection chamber, the larger the free surface area, and the easier it is to generate air entrapment on the wavefront. The lower the filling rate, the lower the liquid level, the smoother the wavefront flow, but a longer slow stroke is required.

[0109] Among them, the larger the inner diameter of the injection chamber, the larger the cross-sectional area of ​​the injection chamber, and the lower the filling rate under the same volume. The inner diameter affects the curvature of the free surface of the molten metal and the surface wave propagation characteristics. The empirical correction coefficient is used to adapt the surface tension and viscosity characteristics of different alloy materials (such as aluminum alloys and magnesium alloys).

[0110] In the above embodiments, when the molten metal is pushed forward by the punch in the injection chamber, its free surface forms a wave-like ripple. If the punch speed is too high, the wavefront will break and entrain gas, causing porosity defects in the casting. Therefore, by setting the wavefront speed, gas entrainment in the molten metal can be prevented. That is, by limiting the slow injection speed below the critical wavefront speed, the free surface of the molten metal is kept stable, preventing wavefront breakage and gas entrainment, thus reducing porosity defects in the casting at the source. The optimal wavefront speed can be obtained for different molds and different casting volumes without the need for repeated manual adjustments. By adjusting the wavefront coefficient, the surface tension and viscosity characteristics of different materials such as aluminum alloys, magnesium alloys, and zinc alloys can be flexibly adapted, providing good versatility. Under the premise of ensuring no gas entrainment, the slow injection speed is increased as much as possible, the low-speed running time is shortened, and the overall production efficiency is improved.

[0111] In this embodiment, for stage 2, a preset speed curve v is determined. set (t) includes:

[0112] Choose a cubic polynomial:

[0113] v(τ1) = C0 + C1τ1 + C2τ1 2 + C3τ1 3 (4)

[0114] Where C0, C1, C2, and C3 are the coefficients of a cubic polynomial, and τ1 is the normalized time independent variable for stage 2, τ1=(t-t1) / T acc1 v(τ1) is the velocity dependent variable corresponding to stage 2;

[0115] Determine the physical boundary conditions for the slow-pressure injection phase, including: starting from rest: when τ1=0, v(0)=0; reaching the wavefront velocity: when τ1=1, v(1)= v gather ; To avoid impact and splashing at the start, the acceleration is 0: when τ1=0, a(0)=0; At the end, smoothly transition to a uniform pushing state without waves, the acceleration is 0: when τ1=1, a(1)=0;

[0116] Substituting the physical boundary conditions of the slow-pressure injection section into the derivative of a cubic polynomial, we obtain the general coefficients of the cubic polynomial, and thus the preset velocity curve v for stage 2. set (t):

[0117] v set (t)= v gather ×(3τ1 2 -2τ1 3 (5)

[0118] Among them, T acc1 It is determined based on the actual displacement x(t) and the sealing safety constraint. The actual displacement x(t) is determined according to the velocity integral formula, expressed as:

[0119] (6)

[0120] The sealing safety constraint is: when there is a time point t s Make the actual displacement x(t) s )=x seal At that time, the current speed v must be strictly satisfied. set (t s )≤v seal .

[0121] In the above embodiments, a cubic polynomial is used to smoothly increase the preset velocity curve from 0 to the wavefront velocity, thereby ensuring a safe correspondence between the time integral and the physical displacement, and the acceleration is constant at the moment of start-up.

[0122] The control process described above for stage 2 employs a cubic polynomial smooth start with a sealing safety constraint during the slow injection phase. This achieves a shock-free transition from rest to wavefront velocity and ensures that the actual speed of the punch reaching the sealing safety position at the injection port does not exceed the safety threshold. This mechanism eliminates fluid splashing and impact surges at the moment of initiation and avoids hydraulic shocks caused by sudden acceleration changes. It provides stable and safe initial flow field conditions for subsequent high-speed injection, significantly improving the operational smoothness of the low-speed injection phase.

[0123] In some embodiments, the triggering of stages 2 to 3 can be when time t reaches t1 + T. acc1 t1 is the end time of phase 1, which is also the start time of phase 2.

[0124] In some embodiments, when time t reaches t1 + T acc1 Then, the entry condition for stage 3 is checked. If the actual displacement x has not reached the ingate, i.e., x ≤ x gate Then, in stage 3, the corresponding control process is executed. Stage 3 involves the injection punch pushing the molten metal at a constant low speed, allowing it to advance slowly and expel air from the cavity. That is, the actual displacement of the injection punch gradually approaches the ingate position until it reaches the ingate, achieving the control objective of stage 3, i.e., when x > x gate If the condition for entering stage 3 is not met, the conditions for entering the next stage will be determined.

[0125] In this embodiment, the preset velocity curve of stage 3 starts from the arrival of the wavefront velocity and continues until the end of stage 3.

[0126] In this embodiment, the actual displacement x(t) reaches the ingate position x gate The time can be t2, in other words, phase 4 is triggered when time t=t2.

[0127] In this embodiment, the position x of the ingate gate The position of the injection punch when it propels the molten metal to the ingate in a rapid phase is represented as:

[0128] x gate =(V biscuit +V runner ) / A chamber (1)

[0129] Among them, A chamber V is the cross-sectional area of ​​the injection chamber containing the molten metal; biscuit V represents the volume of the casting spool; runner This represents the volume of the horizontal runner.

[0130] In this embodiment, the volume of the casting sprue can be the volume of solidified metal remaining in the injection chamber after injection, equivalent to the piston residual section, used to transmit the final pressurization pressure; the volume of the sprue can be the channel volume connecting the injection chamber and the ingate, i.e., the necessary channel for molten metal to enter the mold cavity from the injection chamber; the cross-sectional area of ​​the injection chamber can be the cross-sectional area of ​​the injection chamber, which determines the volume of molten metal corresponding to a unit displacement. Among them, the dimensions of the sprue and sprue are different for different molds, and different molds can be adapted by the above formula (1), without the need for manual estimation of the trigger position.

[0131] In the above embodiments, the acceleration phase is triggered at the instant the molten metal reaches the ingate, which avoids premature acceleration that could cause air entrapment and delayed acceleration that could affect the filling efficiency. This ensures that the molten metal gains sufficient kinetic energy when it enters the ingate, and fills the cavity quickly at an ideal speed, avoiding defects such as cold shuts and flow marks. Based on geometric and volume parameters, the process can be recalculated according to the current mold data before each injection, resulting in good process consistency.

[0132] In some embodiments, stage 4 is an acceleration stage, which can be a stage in which the injection punch rapidly rises from the wavefront velocity to the fast injection velocity, so that the molten metal can obtain enough kinetic energy to quickly fill the cavity.

[0133] In this embodiment, the condition for setting stage 4 can be whether time t is less than or equal to t2 + T. acc2 That is, when t≤t2+T acc2 In this case, proceed to stage 4 and continue execution: use a fifth-order polynomial to change the velocity from the wavefront velocity v gather Smoothly and rapidly pull up to the high injection velocity v fast By forcing continuous velocity, acceleration, and jerk at the start and end points, the water hammer effect caused by sudden velocity changes is prevented. The acceleration duration T during the acceleration phase... acc2 Phase 5 is triggered after the event ends.

[0134] In this embodiment, T acc2 This represents the total duration of Phase 4, i.e., the duration of the acceleration phase. Phase 4 begins at the end time t2 of Phase 3 and lasts for T seconds. acc2 It ends after a certain duration.

[0135] In this embodiment, the fast injection speed v fast The speed set to ensure the molten metal fills the mold cavity properly during the two rapid phases of the injection plunger's propulsion is expressed as follows, based on the filling energy equivalent equation:

[0136] v fast =[V total / (A gate ×t fill )]×(A gate / Achamber (3)

[0137] Among them, V total A represents the total volume of the casting; gate t is the total cross-sectional area of ​​the ingate. fill The target filling time is defined as follows: The total volume of the casting can be the sum of the volumes of the casting body, slag pot, overflow channel, and venting channel; the total cross-sectional area of ​​the ingate can be the sum of the cross-sectional areas of all inlets from the sprue into the mold cavity; the target filling time can be the time required for the cavity to be filled as set by the process.

[0138] In the above embodiments, by calculating and setting a fast injection speed, it is ensured that the molten metal completely fills the mold cavity within the set target filling time, avoiding under-casting or cold shut due to excessively low speed. By calculating the speed in reverse by setting the target filling time, a precise match between speed and casting quality requirements is achieved.

[0139] In some embodiments, the triggering condition for stage 5 is that t reaches t2 + T. acc2 Stage 5 can be a rapid stage, used to control the injection plunger to move at a set high speed and constant motion, rapidly filling the entire mold cavity with molten metal. The condition for entering stage 5 can be that the injection plunger has not exceeded the set deceleration safety position x. safe Furthermore, the calculated differential rate of change did not exceed the precursor threshold ε. pre That is, x < x safe And the differential rate of change < ε pre Then, the control process of stage 5 continues to be executed until the injection punch breaks through the set deceleration safety position and the calculated differential rate of change exceeds the precursor threshold.

[0140] In this embodiment, the deceleration safety position and the warning threshold can be values ​​set for stage 5, and can be customized according to the actual scenario or needs. This disclosure does not limit this.

[0141] In this embodiment, for stage 4, a preset speed curve v is determined. set (t) includes: selecting a fifth-degree polynomial:

[0142] v(τ2)=C0+C1τ2+C2τ2 2 + C3τ2 3 + C4τ2 4 + C5τ2 5 (7)

[0143] Where C0, C1, C2, C3, C4, and C5 are the coefficients of the fifth-order polynomial, and τ2 is the normalized time independent variable for stage 4, τ2=(t-t2) / T acc2 v(τ2) is the velocity dependent variable corresponding to stage 4;

[0144] Determine the physical boundary conditions for the acceleration phase, including: the starting point of the acceleration phase connects to the wavefront velocity of the fast phase: when τ2=0, v(0)= v gather When the target rapid injection velocity is reached at the endpoint: τ2=1, v(1)= v fast ; No impact at start-up, acceleration is 0: when τ2=0, a(0)=0; No overshoot oscillation before entering the fast injection stage, acceleration is 0: when τ2=1, a(1)=0; Smooth torque at the moment the valve core opens, jerk is 0: when τ2=0, j(0)=0; Smooth torque at the moment the valve core stops, jerk is 0: when τ2=1, j(1)=0;

[0145] Substituting the physical boundary conditions of the acceleration section into the derivative of the fifth-order polynomial, we can obtain the coefficients of the general formula of the fifth-order polynomial, and thus obtain the preset velocity curve v of stage 4. set (t):

[0146] v set (t) = v gather +( v fast - v gather )×(10τ2 3 -15τ2 4 +6τ2 5 (8)

[0147] Among them, T acc2 It is determined using a dynamic adaptive criterion based on the target valve opening ratio, expressed as:

[0148] (9)

[0149] In the formula, T valve_step v is the full-scale step response time of the system's servo cartridge valve; max k is the maximum design speed of the injection system. acc A safety tracking factor is set between 1.2 and 1.5; T min The minimum acceleration time limit to prevent high-frequency excitation.

[0150] In the above embodiments, a fifth-order polynomial satisfying six boundary conditions (position, velocity, acceleration, and jerk) is used during the acceleration phase to achieve a globally continuous and smooth increase from the wavefront velocity to the fast injection velocity. Simultaneously, a dynamic acceleration duration adaptive criterion based on the valve core response limit is introduced to ensure that the target velocity change rate does not exceed the physical tracking capability of the servo valve. This eliminates command saturation and high-frequency water hammer excitation, achieving the goal of extreme acceleration without loss of control, and significantly improving the response speed and control accuracy of the fast injection phase.

[0151] In some embodiments, if the entry condition of stage 5 is not met, stage 6 is triggered, i.e., when the injection plunger exceeds the set deceleration safety position x. safe And the differential rate of change exceeds the precursor threshold ε pre That is, phase 6 is triggered when t=t3. Here, t3 can be the end time of phase 5, which is also the start time of phase 6.

[0152] In this embodiment, when the injection punch exceeds the set deceleration safety position x safe And the differential rate of change exceeds the precursor threshold ε pre That is, at t=t3, stage 6 is triggered, which includes: during the operation of stage 5, reading the total disturbance estimate z2 estimated by the autonomous disturbance rejection extended state observer in real time; physically mapping the total disturbance estimate z2 to the comprehensive operating resistance of the injection process; performing a first-order low-pass filter on the total disturbance estimate z2 and then performing differential approximation to extract the comprehensive disturbance drastic rate. When the actual displacement x(t) > x safe And detected >ε pre When the molten metal has filled the main cavity of the mold, it is determined that the molten metal has filled the main cavity of the mold, triggering the entry into stage 6.

[0153] The triggering of stage 6 employs a dual-condition AND gate triggering mechanism: condition 1 is that the injection plunger has entered the deceleration safety zone to avoid premature triggering; condition 2 is that the differential rate of change exceeds the precursor threshold, meaning that the flow resistance increases sharply after the molten metal fills the main cavity. When both conditions are met simultaneously, the system determines that the molten metal has filled the main cavity and actively triggers the deceleration stage.

[0154] The magnitude of the total disturbance estimate z2 reflects the resistance to the flow of molten metal in the mold cavity. Under normal circumstances, the molten metal flows smoothly in the mold cavity with low and gradual flow resistance, resulting in a low value and stable fluctuation. At the moment of filling, the molten metal begins to enter narrow areas such as the slag pot and overflow channel, where the flow resistance increases sharply, causing the value to rise rapidly.

[0155] Specifically, applying a first-order low-pass filter to z2 can eliminate sensor noise and high-frequency interference, preventing false triggering. Furthermore, by using differential approximation, the rate of change of the disturbance over time can be calculated, reflecting the drastic change in flow resistance, thus obtaining the comprehensive disturbance variability rate. .

[0156] In the above embodiments, the deceleration triggering logic extracts the total disturbance estimate and calculates its rate of change in real time through the expanded state observer. Combined with the double-condition AND gate judgment when the punch position exceeds the deceleration safety position, it accurately captures the physical moment when the flow resistance increases sharply after the molten metal fills the main cavity, realizing adaptive braking triggering that decelerates as soon as it is filled, avoiding the undercasting or flash problems caused by traditional fixed position triggering.

[0157] In some embodiments, stage 6 can be a deceleration stage, i.e., a stage in which the injection punch gradually decelerates from a fast injection speed to a safe collision speed to prevent the punch from violently colliding with the mold. The entry condition for stage 6 can be t > t3, up to t ≤ t4.

[0158] In this embodiment, stage 6 can start at time t3 and execute the following control process: using a fifth-order polynomial with a speed lower limit bias to apply tolerance braking to the preset speed curve, reducing the preset speed curve from fast injection speed to a safe collision speed, and when t reaches t3 + T dec Afterwards, maintain a safe collision speed v safe Phase 7 is triggered when t4 is reached, T dec This represents the total duration of the deceleration phase.

[0159] Among them, T dec The duration of stage 6 can be a pre-set duration, which can be customized according to the actual scenario or needs, and this disclosure does not limit it.

[0160] In this embodiment, the deceleration time is calculated by integral matching based on the total volume of the mold overflow groove and the exhaust channel, ensuring that the braking displacement strictly corresponds to the mold tolerance space. After deceleration, v is maintained. safe It runs at a constant speed until the boost trigger moment, that is, until time reaches t4.

[0161] In this embodiment, for stage 6, a preset speed curve v is determined. set (t) includes: selecting a fifth-degree polynomial:

[0162] v(τ3)=C0+C1τ3+C2τ3 2 + C3τ3 3 + C4τ3 4 + C5τ3 5 (10)

[0163] Where C0, C1, C2, C3, C4, and C5 are the coefficients of the fifth-order polynomial, and τ3 is the normalized time independent variable for stage 6, τ3=(t-t3) / T dec v(τ3) is the velocity dependent variable corresponding to stage 4;

[0164] Determine the physical boundary conditions of the deceleration section, including: the initial seam connection speed of the fast injection: when τ3=0, v(0)= v fast When the safe collision speed is reached at the endpoint: τ3=1, v(1)= v safe ;

[0165] The starting point transitions smoothly without sudden water hammer impact, and the acceleration is 0: when τ3=0, a(0)=0;

[0166] The braking action is completed smoothly at the end point, and the acceleration is 0: when τ3=1, a(1)= 0;

[0167] The torque is smooth at the moment the valve core begins to close, and the jerk is 0: when τ3=0, j(0)=0;

[0168] The torque is smooth at the moment the valve core closes, and the jerk is 0: when τ3=1, j(1)=0;

[0169] Substituting the physical boundary conditions of the deceleration section into a fifth-order polynomial yields the general coefficients of the fifth-order polynomial, thus obtaining the preset velocity curve v for stage 6. set (t):

[0170] v set (t) = v safe +( v fast - v safe )×(1-(10τ3 3 -15τ3 4 +6τ3 5 )) (11)

[0171] Among them, T dec Rigidity calculation is performed using a braking integral matching criterion based on mold geometric tolerance: the total volume of the mold overflow channel and venting channel is obtained, and the equivalent safe braking displacement S is calculated in reverse. rem According to the preset velocity curve v of stage 6 set The integral of (t) is obtained from the fact that the displacement is always equal to 0.5(v). fast + v safe )×T dec Based on the characteristics, we obtain:

[0172] T dec =2×S rem / ( v fast + v safe (12)

[0173] After deceleration in stage 6, the preset speed curve v set (t) Maintain a safe collision speed v safe Up to stage 7.

[0174] In order to ensure the end filling pressure holding kinetic energy and eliminate water hammer impact, the preset speed curve adopts a fifth-order polynomial tolerance braking curve with speed lower limit offset. Based on the physical boundary conditions of the deceleration section, the general formula coefficients of the fifth-order polynomial are solved, and finally the fifth-order polynomial expression is obtained.

[0175] In the above embodiment, the speed curve during the deceleration phase adopts a fifth-order polynomial with a lower speed limit bias. Six boundary conditions ensure that the speed, acceleration, and jerk are continuous throughout the process. The deceleration duration is calculated rigidly by integral matching based on the total volume of the mold overflow groove and exhaust channel, so as to achieve a strict correspondence between the braking displacement and the mold tolerance space, thereby achieving a smooth braking effect that is impact-free, prevents flash, prevents under-casting, and is energy-efficient.

[0176] In this embodiment, the preset velocity curve of stage 6 starts from v fast Reduce to v safe The time is t3 + T dec When v is reached safe Then, maintain this speed until t4 is reached, at which point the control process of stage 6 ends and stage 7 is triggered.

[0177] In some embodiments, stage 7 can be a pressurization stage, in which the injection punch stops advancing and the pressurization cylinder establishes high pressure to replenish the solidified molten metal, eliminating internal defects such as shrinkage cavities and porosity.

[0178] In this embodiment, the preset speed curve is reset to zero, and the injection oil inlet valve group is closed. The booster oil inlet cartridge valve is fully opened, and the booster accumulator releases pressure to push the booster cylinder piston, applying high pressure for forced compensation during the solidification and contraction of the molten metal. At the same time, the oil outlet valve switches to the micro-motion compensation and clearance mode, providing an oil discharge channel for the small compensation and displacement of the punch.

[0179] Step S3, based on the actual speed v act A combined feedforward and active disturbance rejection control strategy is employed to obtain the high-frequency disturbance rejection compensation velocity Δv; adaptive planning is used to determine the target back pressure P. 2_target Combined with the target back pressure P 2_target And the actual pressure, based on the preset speed curve v set (t) and compensation speed Δv are used to determine the valve opening command of at least one valve in the die-casting machine in the low-frequency domain and high-frequency domain, respectively, through the inverse physics model.

[0180] In some embodiments, S3 includes: in the feedforward solution branch, the preset velocity curve v set (t) is directly input into the inverse physics model, and the target back pressure P is... 2_target Substituting the actual pressure into the inverse physics model according to different process stages and valve differences, we calculate the low-frequency basic valve opening command u that satisfies macroscopic trajectory tracking and includes the expected macroscopic damping back pressure. base Low-frequency basic valve opening command u base Including the low-frequency basic valve opening command u on the oil inlet side in_base Oil outlet side low-frequency basic valve opening command u out_base In the active disturbance rejection solution branch, the preset velocity curve v is used. set(t) and actual velocity v act A comparison is performed to generate a velocity error, which is then input into the active disturbance rejection regulator in the virtual command domain. The total disturbance estimate is then calculated in real time using an extended state observer, and combined with a linear error feedback control law, the high-frequency disturbance rejection compensation velocity Δv is calculated. Δv is then input into the inverse physics model, and the target back pressure P is calculated. 2_target Substituting the actual pressure into the inverse physics model according to different process stages and different valves, the high-frequency compensated valve opening command u is calculated. comp High-frequency compensated valve opening command u comp Including the high-frequency compensation valve opening command u on the oil inlet side in_comp Oil outlet side high-frequency compensation valve opening command u out_comp Among them, the target back pressure P 2_target The actual pressure is substituted into the inverse physics model according to different process stages and different valves, including: calculating the low-frequency basic valve opening command u on the oil side in the slow stage and the fast stage. out_base Oil outlet side high-frequency compensation valve opening command u out_comp At that time, the target back pressure P 2_target Substituting into the inverse physics model; during the acceleration and second rapid phases, the low-frequency basic valve opening command u on the oil side is calculated. out_base At that time, the target back pressure P 2_target Substitute the actual pressure into the inverse physics model; for the rest, substitute the actual pressure into the inverse physics model.

[0181] S3 can be a composite control strategy, which involves using Active Disturbance Rejection Reduction (ADRC) to obtain a compensated speed within the virtual command domain based on the actual speed of the injection punch. This compensated speed is then combined with the target back pressure from adaptive planning and the acquired actual pressure signal. A preset speed curve and the compensated speed are input into the inverse physics model to finally calculate the initial servo control signal for the injection inlet valve group and the injection outlet valve group. The initial servo control signal includes valve opening commands for at least one valve in the die-casting machine in both the low-frequency and high-frequency domains.

[0182] In this embodiment, the composite control strategy in S3 is implemented through two parallel branches, and the output initial servo control signal can be directionally allocated according to the process stage in step S4. The two parallel branches include a feedforward calculation branch and an active disturbance rejection calculation branch. The feedforward calculation branch is used to calculate the speed curve v according to the preset speed curve v. set (t), target back pressure, and actual pressure are used to determine the valve opening command in the low-frequency domain; the active disturbance rejection branch determines the valve opening command in the high-frequency domain based on the speed error compensation speed Δv, target back pressure, and actual pressure.

[0183] In some embodiments, the feedforward calculation branch may perform the following process to obtain the output low-frequency domain valve opening command: directly inputting a preset velocity curve into the inverse physics model; inputting the target back pressure and actual pressure into the inverse physics model according to different process stages and different valve differences; calculating the low-frequency domain valve opening command u that satisfies macroscopic trajectory tracking and includes the expected macroscopic damping back pressure. base Specifically, for the oil inlet side, the low-frequency basic valve opening command u on the oil inlet side can be calculated. in_base For the oil outlet side, the low-frequency basic valve opening command u on the oil outlet side can be calculated. out_base .

[0184] In this embodiment, the values ​​substituted into the inverse physics model can be different for different stages. For example, for the slow stage and the first fast stage, the target back pressure and the actual pressure can be substituted; for the acceleration stage, the fast stage (or the second fast stage), and the deceleration stage, the target back pressure is substituted; and for the energy storage stage and the pressurization stage, the actual pressure is substituted.

[0185] In the above embodiments, the feedforward calculation branch can ensure that the injection punch moves in accordance with the general trend of the preset speed curve; by introducing the target back pressure, the desired back pressure is established on the oil outlet side, thereby improving the system stiffness.

[0186] In this embodiment, the active disturbance rejection (ADRROC) calculation branch can perform the following process to obtain the high-frequency compensation valve opening command: compare the preset speed curve with the actual speed to generate a speed error; input the speed error into the ADRROC regulator in the virtual command domain, estimate the total disturbance value in real time through the extended state observer, and calculate the high-frequency disturbance rejection compensation speed Δv by combining it with the linear error feedback control law; input Δv into the inverse physics model, and calculate the target back pressure P. 2_target The actual pressure is substituted into the inverse physics model according to different process stages and different valves to finally calculate the high-frequency compensation valve opening command. Specifically, for the oil inlet side, the high-frequency compensation valve opening command for the oil inlet side can be calculated; for the oil outlet side, the high-frequency compensation valve opening command for the oil outlet side can be calculated.

[0187] The parameters used in the inverse physics model can vary depending on the process stage. For example, for the slow and fast stages, the target back pressure and actual pressure can be used; for the acceleration, fast (or second fast) and deceleration stages, the target back pressure can be used; and for the energy storage and pressurization stages, the actual pressure can be used.

[0188] In the above embodiments, different parameters are used for different stages. For the slow stage and the first fast stage, since it is necessary to actively build up high back pressure to prevent creeping oscillations, the actual back pressure may be insufficient. Therefore, the target back pressure and the actual pressure are used. For the acceleration stage, the fast stage (or the second fast stage), and the deceleration stage, since the macroscopic damping is guaranteed by the target back pressure, the high-frequency compensation needs to sense the real pressure to accurately offset the disturbance. Therefore, the target back pressure is used. For the energy storage stage and the pressurization stage, since pressure is no longer actively built up and flexible braking is achieved by passive braking back pressure, the actual pressure is used.

[0189] In the above embodiments, the low-frequency domain and the high-frequency domain can be assigned to different valves. For example, the valve opening command in the low-frequency domain can be assigned to a large-diameter valve to provide a large flow rate, while the valve opening command in the high-frequency domain can be assigned to a small-diameter valve to respond quickly to disturbances.

[0190] In the above embodiments, a dual-path parallel composite control architecture consisting of a feedforward calculation branch and an active disturbance rejection (ADRC) calculation branch is constructed. Combined with differentiated substitution rules for target back pressure and actual pressure, this achieves physical separation between low-frequency trajectory tracking and high-frequency disturbance suppression. The feedforward branch is responsible for macroscopic velocity tracking and back pressure establishment, while the ADRC branch is responsible for real-time cancellation of unmodeled friction and nonlinear filling resistance. This resolves the logical conflict in traditional control where velocity tracking and pressure establishment occur on the same valve core, significantly improving velocity tracking accuracy and disturbance rejection capability across all operating conditions.

[0191] In some embodiments, the low-frequency basic valve opening command u base and high-frequency compensated valve opening command u comp The inverse physics model is obtained through the following steps:

[0192] Obtaining the inverse physics model:

[0193] (17)

[0194] In the formula, u represents the valve opening command, including the low-frequency basic valve opening command u on the oil inlet side. in_base Oil outlet side low-frequency basic valve opening command u out_base , Inlet side high-frequency compensation valve opening command u in_comp Oil outlet side high-frequency compensation valve opening command u out_comp A represents the corresponding hydraulic cylinder working area; V cmd Substitute the target speed; C d K is the flow coefficient; v ρ is the valve's rated flow area gradient; ρ is the hydraulic oil density; ΔP is the driving pressure difference.

[0195] Determine the variable substitution rule: For the effective area A=A1 on the injection inlet side, the driving pressure difference ΔP=Ps -P1 holds true at all times, where P s P1 is the pressure of the injection accumulator, and P2 is the pressure of the inlet chamber. For the effective area of ​​the injection outlet side, A=A2, the driving pressure difference ΔP=P 2_ref –P t , where P t Let P be the oil return pressure from the oil tank. t =0; Target back pressure P on the oil outlet side 2_target The oil outlet pressure P2 is substituted into the inverse physics model according to different usage scenarios; the low-frequency basic valve opening command u on the oil inlet side is calculated. in_base Oil outlet side low-frequency basic valve opening command u out_base At that time, V cmd Substitute v set (t), calculate the opening command u of the high-frequency compensation valve on the oil inlet side. in_comp Oil outlet side high-frequency compensation valve opening command u out_comp At that time, V cmd Substitute Δv; according to the inverse physics model and variable substitution rules, the target back pressure P is... 2_target The directional input is fed into the inverse physics model to replace the oil outlet pressure P2, and the low-frequency basic valve opening command u is calculated. base and high-frequency compensated valve opening command u comp .

[0196] In order to convert the target speed into valve opening command, a unified inverse physics model can be established to clarify the different driving pressure difference calculation methods on the oil inlet and outlet sides; the rules for taking the target input speed and the rules for inputting the target back pressure can be specified to solve the low-frequency basic valve opening command and the high-frequency compensated valve opening command.

[0197] In this embodiment, the variable substitution rules under different conditions may include: for the substitution rule on the oil inlet side, the driving pressure difference uses the actual measured value and does not involve the target back pressure, that is, this calculation method is always valid and applicable to all stages and all branches; for the substitution rule on the oil outlet side, there are two possibilities for the upstream pressure in the driving pressure difference on the right side of the outlet, namely, the target back pressure comes from the dynamic back pressure planning module, and the actual pressure of the oil outlet chamber comes from the oil outlet chamber pressure sensor. The specific choice can be determined according to different usage scenarios (process stage, branch type).

[0198] In this embodiment, the rule for the target input speed is as follows: calculate the opening command of the low-frequency basic valve on the oil inlet side and the opening command of the low-frequency basic valve on the oil outlet side, and input the preset speed; calculate the opening command of the high-frequency compensation valve on the oil inlet side and the opening command of the high-frequency compensation valve on the oil outlet side, and input the compensation speed.

[0199] In this embodiment, the directional input of the target back pressure can be achieved during stages where back pressure needs to be actively established (e.g., slow stage, fast stage). Instead of using the actual value measured by the outlet chamber pressure sensor, the controller uses the target back pressure generated by the dynamic back pressure planning module in the inverse physics model to replace the outlet chamber pressure P2. The calculated outlet valve opening command will cause the outlet valve to actively throttle, establishing the desired back pressure value in the outlet chamber.

[0200] In the above embodiments, by establishing a unified mathematical model for throttling inverse kinematics and clarifying the variable rules for substituting the constant pressure difference on the inlet side, the target back pressure on the outlet side, and the actual pressure differences, high-precision, low-delay mapping of speed commands to valve opening is achieved. This method enables both feedforward control and ADRC compensation to be executed efficiently under the same physical model. Simultaneously, it provides personalized pressure substitution strategies for different stages and control branches, achieving complete decoupling of speed and pressure, and significantly improving the system's control consistency and engineering feasibility.

[0201] In the above embodiments, different generation criteria can be set according to different process stages for the target back pressure.

[0202] In some embodiments, in stage 3, the minimum damping back pressure lower limit is derived based on the sudden change in static and dynamic frictional force of the system, and the maximum allowable back pressure upper limit is derived by back-calculation based on the linear dead zone safety opening constraint of the oil outlet valve port; within the effective physical boundary formed by the minimum damping back pressure lower limit and the maximum allowable back pressure upper limit, the static constant high pressure is selected as the target back pressure of stage 3.

[0203] In this embodiment, for the fast phase, a static constant high-level back pressure is selected based on the constraints of the friction dead zone and the valve linear zone.

[0204] Among them, the static and dynamic frictional force change disturbance can be the difference between static frictional force and dynamic frictional force when the injection punch changes from stationary to moving; the minimum damping back pressure lower limit can be the minimum back pressure requirement to overcome the frictional force change and prevent creep, and can be derived from the frictional force change value; the valve port linear dead zone safety opening can be the minimum opening that needs to be avoided when the servo valve has a nonlinear dead zone in the small opening region; the maximum allowable back pressure upper limit can be the highest back pressure allowed to ensure that the valve core works in the linear region; the static constant high pressure can be a fixed high back pressure value selected between the upper and lower limits.

[0205] In the above embodiment, by setting the stage 3, the low-frequency basic valve opening command and the high-frequency compensation valve opening command of the injection oil valve group can be determined, so as to ensure that the servo valve works in the high-precision linear region, while improving the low-speed motion stiffness and completely suppressing the creeping oscillation.

[0206] In some embodiments, in stage 5, the target back pressure lower limit is derived based on the critical damping criterion for suppressing velocity vibration; a constant low pressure equal to or higher than the target back pressure lower limit is selected as the target back pressure of stage 5.

[0207] In this embodiment, for the rapid phase, a static constant low back pressure can be selected based on the critical vibration absorption damping criterion.

[0208] Among them, the critical damping criterion can be the minimum damping value required to suppress high-frequency velocity vibration; the target back pressure lower limit can be the minimum back pressure required to ensure smoothing of high-frequency jitter; and the constant low pressure can be a low back pressure value that is equal to or slightly higher than the lower limit.

[0209] In some embodiments, in stages 2 and 4, the target back pressure is prohibited from step switching, and a smooth mapping that is forcibly bound to a preset velocity curve is performed: wherein, in stage 2, the preset velocity curve of stage 2 is reused to smoothly establish the target back pressure of stage 2; in stage 4, the preset velocity curve of stage 4 is reused, and an absolutely smooth dynamic unloading mapping is performed from a static constant high pressure to a constant low pressure to determine the target back pressure of stage 4.

[0210] In this embodiment, the slow and acceleration phases can be forcibly bound to a preset speed curve to achieve a smooth transition.

[0211] Specifically, for stage 2, the target back pressure can smoothly transition from the initial back pressure to the high back pressure by reusing the cubic polynomial time normalization function of stage 2, changing synchronously with the velocity curve, thus avoiding shocks caused by sudden changes in back pressure.

[0212] Specifically, for stage 4, the target back pressure reuses the fifth-order polynomial time-normalized function of stage 4, performing an absolutely smooth dynamic unloading mapping from the high back pressure of stage 3 to the low back pressure of stage 5. Thus, during acceleration, the back pressure decreases from the high position of the casting to the low position, and the fifth-order polynomial ensures the continuity of speed, acceleration, and jerk, with no impact during the unloading process; synchronized with the speed curve, it achieves coordinated control of speed increase and back pressure decrease.

[0213] In some embodiments, in stage 6, the active planning constraint of the target back pressure is released, the target back pressure is switched to the oil outlet chamber pressure P2 and brought into the inverse physics model, the oil outlet side valve is controlled to strictly reduce the oil discharge opening in accordance with the preset speed command of the deceleration, and the inertial kinetic energy of the system is absorbed by the valve port throttling damping, so that the oil outlet chamber passively establishes a high-level braking back pressure as the target back pressure.

[0214] In this embodiment, for the deceleration phase, active planning can be deactivated to switch to passive braking back pressure.

[0215] During the deceleration phase, the system needs to quickly dissipate the inertial kinetic energy of the plunger to achieve braking. Therefore, the oil outlet valve can naturally reduce its opening according to the deceleration speed command; the pressure in the oil outlet chamber passively increases (the fluid is compressed); and the throttling effect of the valve port is used to absorb kinetic energy.

[0216] In the above embodiments, dynamic adaptive control of back pressure is achieved by planning differentiated and smoothly transitioning target back pressures for different stages such as slow speed, rapid acceleration, fast speed, and deceleration. High back pressure is used to eliminate creep during the slow speed stage, low back pressure is used to reduce energy loss during the fast speed stage, and the actual pressure is actively switched to achieve flexible braking during the deceleration stage. This solves the speed jitter and energy waste problems caused by abrupt or abrupt back pressure switching in traditional injection systems, significantly improving low-speed stability, high-speed responsiveness, and system energy efficiency.

[0217] In some embodiments, the calculation of the high-frequency disturbance rejection compensation velocity Δv includes: relying on an extended state observer, based on the actual velocity v act Based on the high-frequency disturbance rejection compensation velocity Δv0 of the previous control cycle, the real-time discrete update equation is determined as follows:

[0218] e obs =z1-v act (13)

[0219] z1 = z1 + h × (z2 - β1 × e) obs +b0×△v0) (14)

[0220] z2 = z2 + h × (-β2 × e) obs (15)

[0221] In the formula, z1 is the observed and tracked value of the actual velocity; e obs β1 is the observation error; h is the discrete sampling step size; β1 and β2 are the observer gain parameters; b0 is the system virtual control gain; the total disturbance estimate z2 reflects the mechanical friction of the injection punch and the nonlinear filling flow resistance of the molten metal in the mold cavity;

[0222] Based on the preset velocity curve v set The tracking error e is calculated using (t) and the observed tracking value z1. v = v set (t)-z1;

[0223] Based on the tracking error, a linear error feedback control law is used to generate a virtual primary control quantity u0=k. p ×e v , where k p For proportional gain;

[0224] Based on the total disturbance estimate z2, feedforward compensation is performed on the virtual primary control variable to calculate the high-frequency disturbance rejection compensation speed Δv:

[0225] △v = (u0-z2) / b0 (16)

[0226] One possible calculation path for determining the high-frequency disturbance rejection compensation speed is as follows: establish an extended state observer to estimate the total system disturbance in real time; use a linear error feedback control law to generate a virtual primary control quantity; introduce the total disturbance estimate for feedforward compensation to obtain the final compensation speed.

[0227] In the above embodiments, by placing the Extended State Observer (LESO) in the virtual command domain and using the actual speed and the compensation speed of the previous cycle as observation inputs, real-time, high-precision estimation of the mechanical friction of the injection punch and the flow resistance of the molten metal filling is achieved. Combined with linear error feedback and feedforward compensation, this method can generate compensation speeds before system disturbances occur, significantly improving the anticipation and robustness of the control, significantly suppressing high-frequency speed glitches in the fast injection phase, and achieving micron-level high-precision speed tracking.

[0228] Step S4: Based on the current process stage, the valve opening command is directed to the corresponding valves in the energy storage control cartridge valve, injection control valve group, and booster oil inlet cartridge valve of the die casting machine, driving the die casting machine to complete the complete injection process. The injection control valve group includes an injection oil inlet valve group and an injection oil outlet valve group; the injection oil inlet valve group includes a large-diameter injection oil inlet cartridge valve and a small-diameter injection oil inlet cartridge valve connected in parallel; the injection oil outlet valve group includes a large-diameter injection oil outlet cartridge valve and a small-diameter injection oil outlet cartridge valve connected in parallel.

[0229] In this embodiment, through signal acquisition in S1, stage division and trajectory planning in S2, and composite control calculation in S3, the valve opening command calculated in S3 can be correctly sent to the corresponding physical valve to realize the actual action of the hydraulic system.

[0230] Specifically, the system can determine which stage (2 to 7) the current state is based on the state machine, and then determine the allocation strategy according to the predefined valve allocation rules for each stage; and send the valve opening command calculated in S3 to the designated valve; each valve acts according to the command to complete processes such as slow speed, fast speed, acceleration, fast speed, deceleration, and pressurization.

[0231] In this embodiment, the hydraulic system includes an energy storage control cartridge valve, an injection control valve assembly, and a booster inlet cartridge valve. The energy storage control cartridge valve includes an injection accumulator energy storage control cartridge valve and a booster accumulator energy storage control cartridge valve; the injection control valve assembly includes an injection inlet large / small diameter valve and an injection outlet large / small diameter valve.

[0232] In some embodiments, S4 includes: in stages 2 and 3, the valve opening commands of the forced injection inlet large-diameter cartridge valve and the forced injection outlet large-diameter cartridge valve are zero; the total command, which is the sum of the inlet-side low-frequency basic valve opening command and the inlet-side high-frequency compensated valve opening command, is allocated to the forced injection inlet small-diameter cartridge valve; simultaneously, the outlet-side low-frequency basic valve opening command u out_base With the high-frequency compensation valve opening command u on the oil outlet side out_comp The superimposed total command is assigned to the injection outlet small-diameter cartridge valve; in stages 4 and 5, the low-frequency basic valve opening command on the inlet side is directionally assigned to the injection inlet large-diameter cartridge valve, and the low-frequency basic valve opening command on the outlet side is assigned to the injection outlet large-diameter cartridge valve. out_base Directional allocation to the large-diameter cartridge valve for injection outlet; simultaneously, directional allocation of the high-frequency compensating valve opening command on the inlet side to the small-diameter cartridge valve for injection outlet, and directing the high-frequency compensating valve opening command on the outlet side... out_comp Directional allocation to the injection outlet small-diameter cartridge valve; in stage 6, maintain the enabled state of the injection inlet large-diameter cartridge valve and the injection outlet large-diameter cartridge valve and continue to receive the inlet side low-frequency basic valve opening command u. in_base Oil outlet side low-frequency basic valve opening command u out_base Strictly following the fifth-order polynomial trajectory of the deceleration phase, smooth flow contraction tolerance braking is implemented until the kinetic energy decays to the set low-speed tailing threshold. At this point, the valve opening commands of the injection inlet large-diameter cartridge valve and the injection outlet large-diameter cartridge valve are forcibly cut off. During this period, the injection inlet small-diameter cartridge valve and the injection outlet small-diameter cartridge valve continue to receive high-frequency compensation valve opening commands from the inlet side and the outlet side to continuously perform high-frequency compensation. Simultaneously, output is sent to the booster inlet cartridge valve. A small opening signal is used to pre-charge the booster cylinder with fluid; in stage 7, the valve opening commands of the large-diameter injection inlet cartridge valve, the large-diameter injection outlet cartridge valve, and the small-diameter injection inlet cartridge valve are forced to be zero; a fully open command is output to the booster inlet cartridge valve to trigger the zero-hysteresis transient establishment of the booster pressure; at the same time, the speed tracking control of the small-diameter injection outlet cartridge valve is forcibly released, causing the small-diameter injection outlet cartridge valve to switch to the micro-motion compensation and clearance mode, and a fully open command is output to the small-diameter injection outlet cartridge valve.

[0233] In this embodiment, since the flow rates are small in stages 2 and 3, the large-diameter valve is difficult to control precisely at small openings, while the small-diameter valve has higher control accuracy in the low-flow-rate region; the inlet side is responsible for speed tracking, and the outlet side is responsible for back pressure establishment. The two are independent, thus completely eliminating internal friction.

[0234] In this embodiment, for stages 4 and 5, since a large flow rate is required, the large-diameter valve provides the macroscopic flow rate basis, and the high-frequency disturbance is compensated by the small-diameter valve with a rapid response, thus realizing frequency domain decoupling of "large valve supplying flow rate and small valve compensating for disturbance".

[0235] In this embodiment, for stage 6, flexible braking is still required. The large-diameter valve reduces its opening according to the deceleration curve, while the small-diameter valve continues to compensate for high-frequency disturbances to ensure smooth deceleration. Taking advantage of this deceleration time window when the injection flow demand drops sharply, the booster cylinder is pre-filled to overcome the hydraulic oil volume compression effect and mechanical assembly clearances.

[0236] In this embodiment, for stage 7, the punch basically stops, all oil inlet valves are closed, and oil supply stops; the booster cartridge valve is fully open, and the booster accumulator releases pressure instantly to achieve zero-hysteresis pressure build-up; the oil outlet valve is fully open to provide an oil discharge channel for the small displacement of the punch during the feeding process, preventing the system from stalling.

[0237] In the above embodiments, by dynamically reorganizing and directionally allocating valve opening commands at different process stages through a state machine, the entire process hardware scheduling is achieved, including independent control of small valves in the slow stage, flow supply from large valves and disturbance compensation from small valves in the fast stage, pre-filling in the deceleration stage, and yielding and shrinkage compensation in the pressurization stage. This allocation logic enables each valve to perform its own function and work collaboratively throughout the entire process, achieving zero-hysteresis pressure building and high-precision shrinkage compensation at the hardware execution level, significantly improving the internal density and molding quality of large die-cast parts.

[0238] Figure 2 This is a schematic diagram of the injection hydraulic system architecture of a die-casting machine according to an embodiment of this disclosure, as shown below. Figure 2 As shown, the system 200 includes a controller 210, a power and energy storage unit 220, an execution unit 230, an injection control valve group 240, a boost control valve group 250, and a status detection unit 260.

[0239] The controller is used to execute the injection control method proposed in this embodiment. The power and energy storage unit includes an oil tank, a hydraulic source, an injection accumulator, a booster accumulator, and an energy storage control cartridge valve. The oil outlet of the hydraulic source is connected to the injection accumulator and the booster accumulator respectively through the energy storage control cartridge valve. The execution unit includes an injection cylinder, an injection punch and a check valve disposed within the injection cylinder, and a booster cylinder. The injection control valve group includes an injection inlet valve group and an injection outlet valve group. The injection inlet valve group includes a large-diameter injection inlet cartridge valve and a small-diameter injection inlet cartridge valve connected in parallel and connected to the injection accumulator. The system includes: an injection accumulator and an injection cylinder; an injection outlet valve assembly consisting of a large-diameter injection outlet cartridge valve and a small-diameter injection outlet cartridge valve connected in parallel, and connected between the injection cylinder outlet and the oil tank; a booster control valve assembly consisting of a booster inlet cartridge valve connected between the booster accumulator and the booster cylinder outlet, with the booster cylinder outlet directly connected to the oil tank; and a status detection unit consisting of an accumulator pressure sensor located at the injection accumulator, an inlet chamber pressure sensor located at the injection cylinder outlet, an outlet chamber pressure sensor located at the outlet, and a displacement sensor connected to the injection punch.

[0240] For example, a schematic diagram of the architecture of a specific embodiment of the injection hydraulic system of a die-casting machine can be shown as follows: Figure 3 As shown, the injection hydraulic system includes a power and energy storage unit, an execution unit, an injection control valve group, a booster control valve group, and a status detection unit. The power and energy storage unit includes an oil tank, a hydraulic power source, an injection accumulator, a booster accumulator, and an energy storage control cartridge valve. The oil outlet of the hydraulic power source is connected to the injection accumulator and the booster accumulator respectively through the energy storage control cartridge valve. The execution unit includes an injection cylinder, an injection punch and a check valve installed in the injection cylinder, and a booster cylinder. The injection control valve group includes an injection inlet valve group and an injection outlet valve group. The injection inlet valve group consists of a large-diameter injection inlet servo valve and a small-diameter injection inlet servo valve connected in parallel. The system comprises an injection accumulator and an injection cylinder inlet valve assembly; an injection outlet valve assembly consisting of a large-diameter injection outlet servo valve and a small-diameter injection outlet servo valve connected in parallel, and connected between the injection cylinder outlet and the oil tank; a booster control valve assembly consisting of a booster inlet cartridge valve connected between the booster accumulator and the booster cylinder inlet valve, with the booster cylinder outlet directly connected to the oil tank; and a status detection unit including an accumulator pressure sensor located at the injection accumulator, an inlet chamber pressure sensor and an outlet chamber pressure sensor located at the injection cylinder inlet and outlet respectively, and a displacement sensor connected to the injection punch.

[0241] Figure 4 This is a control flowchart of an injection system for a die-casting machine according to an embodiment of this disclosure. Specifically, Figure 4The pressure injection control process is illustrated, including system signal acquisition, multi-stage state division, and target curve planning.

[0242] The entry condition for the energy storage stage is t≤t1, at which point the energy storage control cartridge valve opens, and the slow stage is triggered when t=t1; the entry condition for the slow stage is t≤t1+T. acc1 During the slow phase, the small servo valve opens until the slow phase lasts for a duration of T. acc1 Triggering a fast phase; the condition for entering a fast phase is x ≤ x. gate Control the opening degree of the small servo valve until x reaches x gate The acceleration phase is triggered, and the condition for entering the acceleration phase is t ≤ t2 + T. acc2 During the acceleration phase, the small servo valve opening is controlled while the large servo valve opens, until the duration of the acceleration phase reaches T. acc2 This triggers the second fast phase; the entry condition for the second fast phase is x < x. safe and <ε pre In the second fast phase, the opening of the small servo valve and the large servo valve are controlled simultaneously until x(t) > x. safe and >ε pre The deceleration phase is triggered when t ≤ t4. During the deceleration phase, the small servo valve is opened and the large servo valve is closed until time t reaches t4, at which point the pressurization phase is triggered. The pressurization phase is triggered when pressurization is completed. During the pressurization phase, the small servo valve for oil outlet is opened and the pressurization cartridge valve is opened until pressurization is completed, at which point the control process ends.

[0243] Figure 5 This is a schematic diagram illustrating the design principle of a target injection speed curve for a die-casting machine, provided according to an embodiment of this disclosure. Specifically, Figure 5 The target injection velocity curve is shown, which is the preset velocity curve in the above embodiment.

[0244] Among them, 0-t4 is the energy storage stage; t1 to t1+T acc1 This is the slow phase; t1 + T acc1 The period up to t2 is a fast phase; the period from t2 to t2+T is a fast phase. acc2 This is the acceleration phase; t2 + T acc2 The period from t3 to t4 is the second rapid phase; the period from t3 to t4 is the deceleration phase; and the period from t4 to the completion of pressurization is the pressurization phase.

[0245] Figure 6 This is a schematic diagram illustrating the decoupling and directional allocation principle of valve commands in an injection system of a die-casting machine, according to an embodiment of this disclosure. Specifically, as shown... Figure 6As shown, the target speed (preset speed curve) is simultaneously fed into the feedforward branch (directly converted to low-frequency basic opening via inverse physics model) and the ADRC branch (active disturbance rejection controller) to generate a compensated speed, which is then converted to a high-frequency compensated opening via inverse physics model. The outputs of the two branches are decoupled and directionally allocated according to the process stage to the large / small diameter valves, thus inputting the final valve control signal into the injection hydraulic system, thereby achieving high-precision composite control of "feedforward for macroscopic control and ADRC for microscopic control". The sensor can collect the actual speed of the injection punch in the system in real time and output the observed speed through the extended disturbance observer. The observed speed can be input to the active disturbance rejection controller along with the target speed to complete the next control process.

[0246] Figure 7 This is a schematic diagram of a liquid nitrogen jet partitioned cooling multi-condition system provided according to an embodiment of this disclosure. Specifically, the valves in the first column are numbered according to... Figure 3 The component number is in the middle; u represents the valve opening command, and its subscript in represents the relevant opening on the injection inlet side, out represents the relevant opening on the injection outlet side, base represents the opening calculated from the preset speed, and comp represents the opening calculated from the compensated speed; P 2_target The target back pressure is substituted into the inverse model representing the calculated opening, while the rest are substituted into the sensor pressure.

[0247] In summary, the technical effects of this disclosure are as follows:

[0248] 1. In terms of injection trajectory planning, the die-casting process requirements and the physical boundaries of the electromechanical system are used as hard constraints, enabling adaptive optimization generation of the target curve. This ensures that each injection curve perfectly matches the current mold characteristics and hardware limits, theoretically eliminating fluid excitation and process mismatch.

[0249] 2. In terms of composite control algorithms and back pressure planning, a differentiated substitution strategy of "feedforward pressure generation and self-disturbance rejection and leakage compensation" is adopted to achieve high-precision smooth control under all operating conditions.

[0250] 3. In terms of hardware valve allocation and process integration, the system utilizes independent control of inlet and outlet oil and a parallel architecture of large and small valves to achieve precise valve scheduling for different process stages. This allows each hardware valve to perform at its maximum efficiency, eliminating the process defects of traditional die-casting machines, such as slow pressure build-up and easy stalling during pressurization, from the root of the physical hardware.

[0251] In this disclosure, unless otherwise stated, elements expressed in the singular form, such as "a," "an," "the," "the," "the," "the," "the," "the," "this," etc., can mean "one and only one," or "one or more," "at least one," etc. For example, when using articles such as "a," "an," "the," etc. in translation, the noun following the article can be understood as either a singular or a plural expression.

[0252] In the embodiments of this disclosure, "multiple" refers to "two or more", "at least two", etc. The terms "at least one of A or B", "at least one of A and B", "one or more", "a plurality of", "multiple" etc. can be used interchangeably.

[0253] In some embodiments, the notation "A or B" may include the following technical solutions, depending on the situation: in some embodiments, A (execute A regardless of whether a branch B exists); in some embodiments, B (execute B regardless of whether a branch A exists); in some embodiments, execution is selected from A and B (A and B are selectively executed). The same applies when there are more branches such as A, B, and C.

[0254] The prefixes "first," "second," etc., used in the embodiments of this disclosure are merely for distinguishing different descriptive objects and do not impose restrictions on the position, order, priority, quantity, or content of the descriptive objects. The description of the descriptive objects is found in the claims or the context of the embodiments, and the use of prefixes should not constitute unnecessary restrictions. For example, if the descriptive object is a "field," the ordinal numbers preceding "field" in "first field" and "second field" do not restrict the position or order of the "fields." "First" and "second" do not restrict whether the "fields" they modify are in the same message, nor do they restrict the order of "first field" and "second field." Similarly, if the descriptive object is a "level," the ordinal numbers preceding "level" in "first level" and "second level" do not restrict the priority between "levels." Furthermore, the number of descriptive objects is not limited by ordinal numbers and can be one or more. For example, in "first device," the number of "devices" can be one or more. Furthermore, the objects modified by different prefixes can be the same or different. For example, if the object being described is "device", then "first device" and "second device" can be the same device or different devices, and their types can be the same or different. Similarly, if the object being described is "information", then "first information" and "second information" can be the same information or different information, and their content can be the same or different.

[0255] In some embodiments, terms such as “in response to…”, “in response to determining…”, “in the case of…”, “when…”, “when…”, “if…”, etc. can be used interchangeably. These descriptions all refer to the device making a corresponding action under certain objective circumstances. They do not necessarily limit the time, nor do they require the device to make a judgment action when implementing it, nor do they mean that there must be other limitations.

[0256] In some embodiments, the terms “greater than,” “greater than or equal to,” “not less than,” “more than,” “more than or equal to,” “not less than,” “higher than,” “higher than or equal to,” “not lower than,” and “above” can be used interchangeably, as can the terms “less than,” “less than or equal to,” “not greater than,” “less than,” “less than or equal to,” “not more than,” “lower than,” “lower than or equal to,” “not higher than,” and “below”.

[0257] In some embodiments, devices, etc., may be interpreted as physical or virtual, and their names are not limited to those described in the embodiments. Terms such as “device,” “equipment,” “device,” “node,” “unit,” “section,” “system,” “entity,” and “body” are interchangeable.

[0258] It should be understood that the specific embodiments described above are merely illustrative or explanatory of the principles of the invention and do not constitute a limitation thereof. Therefore, any modifications, equivalent substitutions, improvements, etc., made without departing from the spirit and scope of the invention should be included within the protection scope of the invention. Furthermore, the appended claims are intended to cover all variations and modifications falling within the scope and boundaries of the appended claims, or equivalent forms of such scope and boundaries.

Claims

1. A method for controlling the injection of a die-casting machine, characterized in that, include: S1: Real-time acquisition of the actual displacement x(t) of the injection punch collected by the displacement sensor of the die-casting machine and the actual pressure collected by the pressure sensor of the die-casting machine, and determination of the actual velocity v of the injection punch based on the actual displacement x(t). act The actual pressure includes the pressure of the injection accumulator, P. s The inlet pressure P1 and the outlet pressure P2; S2: Based on the actual displacement x(t), the injection process is divided into multiple consecutive process stages, and a preset speed curve v of the injection punch is determined for each process stage. set (t), where t is the injection time; S3: Based on the actual speed v act A combined feedforward and active disturbance rejection control strategy is employed to obtain the high-frequency disturbance rejection compensation velocity Δv; adaptive planning is used to determine the target back pressure P. 2_target Combined with the target back pressure P 2_target And the actual pressure, according to the preset speed curve v set (t) and the compensation speed Δv, through the inverse physics model, determine the valve opening command of at least one valve in the die-casting machine in the low-frequency domain and the high-frequency domain respectively; S4: Based on the current process stage, the valve opening command is directed to the corresponding valves in the energy storage control cartridge valve, injection control valve group, and booster oil inlet cartridge valve of the die casting machine, driving the die casting machine to complete the complete injection process. The injection control valve group includes an injection oil inlet valve group and an injection oil outlet valve group; the injection oil inlet valve group includes a large-diameter injection oil inlet cartridge valve and a small-diameter injection oil inlet cartridge valve connected in parallel; the injection oil outlet valve group includes a large-diameter injection oil outlet cartridge valve and a small-diameter injection oil outlet cartridge valve connected in parallel.

2. The method according to claim 1, characterized in that, The multiple consecutive process stages include: Phase 1 is the energy storage phase, used to store energy in the injection accumulator and booster accumulator of the die-casting machine. In this energy storage phase, v set (t)=0; when time t reaches the required energy storage duration, i.e., t=t1, stage 2 is triggered; Phase 2 is the slow phase, in which a cubic polynomial is used to transform the preset speed curve v set (t) smoothly increases from 0 to the wavefront velocity v gather By applying a sealing safety constraint, the injection punch is ensured to reach the set sealing safety position x at the soup inlet. seal That is, t=t s At that time, the actual speed v act Not greater than the set sealing safety speed v seal When t reaches t1 + T acc1 Time-triggered phase 3, t1 is the start time of the slow phase; T acc1 The total duration of the slow phase; Phase 3 is a fast phase, when the preset speed curve v set (t) reaches the wavefront velocity v gather Then, with the v gather Constantly push out molten metal to release air; when the actual displacement x(t) reaches the ingate position x gate That is, stage 4 is triggered when t=t2; Phase 4 is the acceleration phase, in which a globally continuous fifth-order polynomial is used to optimize the preset velocity curve v. set (t) from v gather Rapidly pull up to high compression velocity v fast This causes the injection punch to rapidly fill the cavity with molten metal; when t reaches t2 + T acc2 After triggering phase 5, t2 is the start time of the acceleration phase, T acc2 The total duration of the acceleration phase; Phase 5 is the rapid phase, and the preset speed curve v set (t) Maintain the fast injection speed v fast Constant; relying on the active disturbance rejection expansion state observer, the total disturbance estimate and the differential rate of change of the total disturbance estimate of the die casting machine during the die casting process are estimated in real time. When the injection punch breaks through the set deceleration safety position x safe And the differential rate of change exceeds the precursor threshold ε pre That is, stage 6 is triggered when t=t3; Stage 6 is the deceleration stage, in which a fifth-order polynomial with a lower speed limit bias is used to modify the preset speed curve v. set (t) Implement tolerance braking, and the preset speed curve v set (t) from the fast injection velocity v fast Reduce to a safe collision speed v safe v safe >0; when t reaches t3+T dec Then, maintain the safe collision speed v safe Phase 7 is triggered when t4 is reached, t3 is the start time of the deceleration phase, and T is the starting time of the deceleration phase. dec The total duration of the deceleration phase; Stage 7 is the pressurization stage, and the preset speed curve v set (t) Forced to zero.

3. The method according to claim 1, characterized in that, The preset velocity curve v set The key spatial nodes and velocity thresholds in (t) include the ingate location x. gate Wavefront velocity v gather High-pressure injection velocity v fast At least one of the following: The location of the ingate x gate The position of the injection plunger when it pushes the molten metal to the ingate in the first rapid phase is represented as: x gate =(V biscuit +V runner ) / A chamber (1) Among them, A chamber V is the cross-sectional area of ​​the injection chamber containing the molten metal; biscuit V represents the volume of the casting spool; runner This refers to the volume of the horizontal runner; The wavefront velocity v gather The speed set by which the injection punch propels the molten metal forward during the rapid phase to prevent air entrapment in the molten metal is expressed, according to the shallow water wave critical velocity theory, as follows: （2） Where f is the injection chamber filling rate; D is the inner diameter of the injection chamber; g is the gravitational acceleration; and k is the wavefront coefficient. The high injection speed v fast The speed set by which the injection punch propels the molten metal forward during the two rapid phases to ensure that the molten metal fills the cavity normally is expressed according to the filling energy equivalent equation as follows: in fast =[V total / (A gate ×t fill )]×(A gate / A chamber ) (3) Among them, V total A represents the total volume of the casting; gate t is the total cross-sectional area of ​​the ingate. fill Time required to fill the target area.

4. The method according to claim 2, characterized in that, For stage 2, the preset speed curve v is determined. set (t) includes: Select the cubic polynomial: v(τ1)=C0+C1τ1+C2τ1 2 + C3τ1 3 (4) Where C0, C1, C2, and C3 are the coefficients of a cubic polynomial, and τ1 is the normalized time independent variable of stage 2, τ1=(t-t1) / T acc1 v(τ1) is the velocity dependent variable corresponding to stage 2; Determine the physical boundary conditions for the slow-pressure injection phase, including: Starting from rest: when τ1=0, v(0)=0; When the wavefront velocity τ1=1 is reached, v(1)= v gather ; To avoid impact and splashing during startup, the acceleration is 0: when τ1=0, a(0)=0; The endpoint smoothly transitions to a uniform pushing state without waves, and the acceleration is 0: when τ1=1, a(1)=0; Substituting the physical boundary conditions of the slow-pressure injection section into the derivative of the cubic polynomial, we obtain the general coefficients of the cubic polynomial, and thus obtain the preset velocity curve v of stage 2. set (t): v set (t)= v gather ×(3τ1 2 -2τ1 3 ) (5) Among them, T acc1 It is determined based on the actual displacement x(t) and the sealing safety constraint. The actual displacement x(t) is determined according to the velocity integral formula, expressed as: （6） The sealing safety constraint is: when there is a time point t s Make the actual displacement x(t) s )=x seal At that time, the current speed v must be strictly satisfied. set (t s )≤v seal .

5. The method according to claim 2, characterized in that, For stage 4, the preset speed curve v is determined. set (t) includes: Select the fifth-degree polynomial: v(τ2)=C0+C1τ2+C2τ2 2 + C3τ2 3 + C4τ2 4 + C5τ2 5 (7) Where C0, C1, C2, C3, C4, and C5 are the coefficients of a fifth-order polynomial, and τ2 is the normalized time independent variable of stage 4, τ2=(t-t2) / T acc2 v(τ2) is the velocity dependent variable corresponding to stage 4; Determine the physical boundary conditions for the acceleration phase, including: The starting point of the acceleration phase connects to the wavefront velocity of the fast phase: when τ2=0, v(0)= v gather ; When the target rapid injection velocity is reached at the endpoint: τ2=1, v(1)= v fast ; There is no impact at the start, and the acceleration is 0: when τ2=0, a(0)= 0; Before entering the fast injection stage, there is no overshoot oscillation, and the acceleration is 0: when τ2=1, a(1)=0; The torque is smooth at the moment the valve core opens, and the jerk is 0: when τ2=0, j(0)=0; The torque is smooth at the instant the valve core stops, and the jerk is 0: when τ2=1, j(1)=0; Substituting the physical boundary conditions of the acceleration section into the derivative of the fifth-order polynomial, the coefficients of the general formula of the fifth-order polynomial can be obtained, thus yielding the preset velocity curve v of stage 4. set (t): v set (t) = v gather +( v fast - v gather ) ×(10τ2 3 -15τ2 4 +6τ2 5 ) (8) Among them, T acc2 It is determined using a dynamic adaptive criterion based on the target valve opening ratio, expressed as: （9） In the formula, T valve_step v is the full-scale step response time of the system's servo cartridge valve; max k is the maximum design speed of the injection system. acc A safety tracking factor is set between 1.2 and 1.5; T min The minimum acceleration time limit to prevent high-frequency excitation.

6. The method according to claim 2, characterized in that, When the injection punch breaks through the set deceleration safety position x safe And the differential rate of change exceeds the precursor threshold ε pre That is, stage 6 is triggered at t=t3, including: During the operation of Phase 5, the total disturbance estimate z2 estimated by the active disturbance rejection extended state observer is read in real time. The total disturbance estimate z2 is physically mapped to the comprehensive operating resistance of the injection process. After applying a first-order low-pass filter to the total disturbance estimate z2, a differential approximation is performed to extract the comprehensive disturbance variability rate. ; When the actual displacement x(t) > x safe And detected >ε pre When it is determined that the molten metal has filled the main cavity of the mold, the process is triggered to enter stage 6.

7. The method according to claim 2, characterized in that, For stage 6, the preset speed curve v is determined. set (t) includes: Select the fifth-degree polynomial: v(τ3)=C0+C1τ3+C2τ3 2 + C3τ3 3 + C4τ3 4 + C5τ3 5 (10) Where C0, C1, C2, C3, C4, and C5 are the coefficients of a fifth-order polynomial, and τ3 is the normalized time independent variable of stage 6, τ3=(t-t3) / T dec v(τ3) is the velocity dependent variable corresponding to stage 4; Determine the physical boundary conditions of the deceleration section, including: When the starting seam is connected to the fast injection velocity: τ3=0, v(0)= v fast ; When the safe collision speed is reached at the endpoint: τ3=1, v(1)= v safe ; The starting point transitions smoothly without sudden water hammer impact, and the acceleration is 0: when τ3=0, a(0)=0; The braking action is completed smoothly at the end point, and the acceleration is 0: when τ3=1, a(1)= 0; The torque is smooth at the moment the valve core begins to close, and the jerk is 0: when τ3=0, j(0)=0; The torque is smooth at the moment the valve core closes, and the jerk is 0: when τ3=1, j(1)=0; Substituting the physical boundary conditions of the deceleration section into the fifth-order polynomial yields the general coefficients of the fifth-order polynomial, thus obtaining the preset velocity curve v of stage 6. set (t): v set (t) = v safe +( v fast - v safe )×(1-(10τ3 3 -15 sq m 4 +6t3 5 )) (11) Among them, T dec Rigidity calculation is performed using a braking integral matching criterion based on mold geometric tolerance: the total volume of the mold overflow channel and venting channel is obtained, and the equivalent safe braking displacement S is calculated in reverse. rem According to the preset speed curve v of stage 6 set The integral of (t) is obtained from the fact that the displacement is always equal to 0.5(v). fast + v safe )×T dec Based on the characteristics, we obtain: T dec =2×S rem / ( v fast + v safe ) (12) After deceleration in stage 6, the preset speed curve v set (t) Maintain the safe collision speed v safe Until stage 7.

8. The method according to claim 1, characterized in that, S3 includes: In the feedforward solution branch, the preset velocity curve v set (t) is directly input into the inverse physics model, and the target back pressure P is... 2_target The actual pressure, differentiated according to different process stages and valves, is substituted into the inverse physics model to calculate the low-frequency basic valve opening command u that satisfies macroscopic trajectory tracking and includes the expected macroscopic damping back pressure. base The low-frequency basic valve opening command u base Including the low-frequency basic valve opening command u on the oil inlet side in_base Oil outlet side low-frequency basic valve opening command u out_base ; In the active disturbance rejection solution branch, the preset velocity curve v set (t) and the actual velocity v act A comparison is performed to generate a velocity error, which is then input into the active disturbance rejection regulator in the virtual command domain. The total disturbance estimate is estimated in real time using the extended state observer, and the high-frequency disturbance rejection compensation velocity Δv is calculated by combining it with a linear error feedback control law. Δv is then input into the inverse physics model, and the target back pressure P is calculated. 2_target The actual pressure, differentiated according to different process stages and valve types, is substituted into the inverse physics model to calculate the high-frequency compensated valve opening command u. comp The high-frequency compensated valve opening command u comp Including the high-frequency compensation valve opening command u on the oil inlet side in_comp Oil outlet side high-frequency compensation valve opening command u out_comp ; Wherein, the target back pressure P 2_target The actual pressure is substituted into the inverse physics model according to different process stages and different valves, including: calculating the low-frequency basic valve opening command u on the oil outlet side during the slow stage and the fast stage. out_base The oil outlet side high-frequency compensation valve opening command u out_comp At that time, the target back pressure P 2_target Substituting into the inverse physics model; during the acceleration phase and the second rapid phase, calculate the low-frequency basic valve opening command u on the oil outlet side. out_base At that time, the target back pressure P 2_target Substitute the values ​​into the inverse physics model; for the rest, substitute the actual pressure into the inverse physics model.

9. The method according to claim 8, characterized in that, The calculation of the high-frequency interference rejection compensation speed Δv includes: Based on the extended state observer, and based on the actual velocity v act Based on the high-frequency disturbance rejection compensation rate Δv0 of the previous control cycle, the real-time discrete update equation is determined as follows: yes obs =z1-v act (13) z1=z1+h · (z2-β1×e obs +b0×△v0) (14) z2=z2+h×(-β2×e obs ) (15) In the formula, z1 is the observed and tracked value of the actual velocity; e obs β1 is the observation error; h is the discrete sampling step size; β1 and β2 are the observer gain parameters; b0 is the system virtual control gain; the total disturbance estimate z2 reflects the mechanical friction of the injection punch and the nonlinear filling flow resistance of the molten metal in the mold cavity; Based on the preset velocity curve v set (t) and the observed tracking value z1 are used to calculate the tracking error e v = v set (t)-z1; Based on the tracking error, the linear error feedback control law is used to generate a virtual primary control quantity u0=k. p ×e v , where k p For proportional gain; Based on the total disturbance estimate z2, feedforward compensation is performed on the virtual primary control variable to calculate the high-frequency disturbance rejection compensation speed Δv: △v= (u0-z2) / b0 (16).

10. The method according to claim 8, characterized in that, The low-frequency basic valve opening command u base and the high-frequency compensated valve opening command u comp The following steps are used to calculate: Obtain the inverse physics model: （17） In the formula, u is the valve opening command, including the low-frequency basic valve opening command u on the oil inlet side. in_base Oil outlet side low-frequency basic valve opening command u out_base , Inlet side high-frequency compensation valve opening command u in_comp Oil outlet side high-frequency compensation valve opening command u out_comp A represents the corresponding hydraulic cylinder working area; V cmd Substitute speed into the target; C d K is the flow coefficient; v ρ is the valve's rated flow area gradient; ρ is the hydraulic oil density; ΔP is the driving pressure difference. Determine the rules for variable substitution: For the effective area A=A1 on the injection inlet side, the driving pressure difference ΔP=P s -P1 holds true at all times, where P s P1 is the pressure of the injection accumulator, and P2 is the pressure of the oil inlet chamber. For the effective area A=A2 on the injection oil outlet side, the driving pressure difference ΔP=P 2_ref –P t , where P t Let P be the return oil pressure from the oil tank. t =0; the target back pressure P is applied to the oil outlet side. 2_target The oil outlet pressure P2 is substituted into the inverse physics model according to different usage scenarios; Calculate the opening command u of the low-frequency basic valve on the oil inlet side. in_base The oil outlet side low-frequency basic valve opening command u out_base At that time, V cmd Substitute v set (t), calculate the opening command u of the high-frequency compensation valve on the oil inlet side. in_comp The oil outlet side high-frequency compensation valve opening command u out_comp At that time, V cmd Substitute it into △v; Based on the inverse physics model and the variable substitution rule, the target back pressure P is... 2_target The directional input is given to the inverse physics model to replace the oil outlet pressure P2, and the low-frequency basic valve opening command u is calculated. base and the high-frequency compensated valve opening command u comp .

Images