Forging method and equipment for automobile oil nozzle forge piece

Abstract

The invention is suitable for the technical field of automobile fuel injection nozzle forging, and particularly relates to an automobile fuel injection nozzle forge piece forging method and equipment. The method comprises the steps that metal flow resistance, metal flow speed, forging vibration data and lateral force are obtained; determining a target flow velocity based on the metal flow resistance, the metal flow velocity and the forging vibration data; determining an axial strain coefficient and a radial strain coefficient based on the metal flow resistance, the metal flow velocity, the forging vibration data and the lateral force; based on the axial strain coefficient and the target flow velocity, generating a sliding block control instruction, and based on the radial strain coefficient, generating an ejector rod control instruction; and according to the sliding block control instruction and the ejector rod control instruction, a forging and pressing device and a hydraulic device are controlled to forge the automobile oil nozzle forge piece. According to the method, the forge piece forming precision, the production efficiency and the intelligence and automation level of the forging process can be remarkably improved.

Description

Technical Field

[0001] The present invention relates to the technical field of automobile fuel injector forging, and in particular to a forging method and equipment for automobile fuel injector forgings. Background Art

[0002] Automotive fuel injectors are responsible for precisely spraying fuel into the engine's combustion chamber, ensuring complete combustion and efficient engine operation. Injector quality directly impacts engine performance, emissions standards, and fuel economy. As the automotive industry's environmental and performance requirements increase, injector design and manufacturing technologies are constantly evolving. Forging, due to its ability to provide higher material strength, excellent fatigue resistance, and superior wear resistance, has become a key technology for manufacturing high-precision injectors.

[0003] In the existing technology, in the traditional forging process of automobile fuel injector forgings, the process path is usually preset as fixed parameters in the process design stage, and is repeated with the same loading path and motion trajectory in actual production. It cannot adapt to the real-time fluctuations of material properties and is prone to cause a series of quality defects including folding, flow marks, insufficient filling, etc. Summary of the Invention

[0004] The embodiments of the present application provide a method and equipment for forging automobile fuel injector forgings, which can solve the problem of insufficient forging forming accuracy during the forging process of automobile fuel injector forgings due to the fact that traditional forging relies on a fixed path and cannot adapt to material fluctuations.

[0005] In a first aspect, an embodiment of the present application provides a method for forging an automobile fuel injector forging, comprising: Obtaining metal flow resistance, metal flow velocity, forging vibration data, and lateral force; wherein the forging vibration data includes die vibration amplitude and die vibration frequency; the lateral force is used to characterize the intensity of metal radial flow; the lateral force includes expansion thrust along the x-axis direction and expansion thrust along the y-axis direction; determining a target flow rate based on the metal flow resistance, the metal flow rate, and the forging vibration data; determining an axial strain factor and a radial strain factor based on the metal flow resistance, the metal flow velocity, the forging vibration data, and the lateral force; Based on the axial strain coefficient and the target flow rate, a slider control instruction is generated, and based on the radial strain coefficient, a push rod control instruction is generated; wherein the slider control instruction is used to control the slider to apply pressure to the metal along the axis of the automobile fuel injector forging to form the forging, and the push rod control instruction is used to adjust the pressure of four push rods to guide the flow of the metal in the radial direction, and the four push rods are all located in the horizontal direction and are respectively distributed in the positive and negative axis directions of the x-axis and the positive and negative axis directions of the y-axis; According to the slider control instruction and the push rod control instruction, the forging device and the hydraulic device are respectively controlled to forge the automobile fuel injection nozzle forging.

[0006] The above technical solutions in the embodiments of the present application have at least the following technical effects: The forging method for an automotive fuel injector forging provided in the present application first obtains metal flow resistance, metal flow velocity, forging vibration data (die vibration amplitude and die vibration frequency), and lateral force (characterizing the strength of radial metal flow, including expansion thrust along the x-axis and expansion thrust along the y-axis). Then, based on the metal flow resistance, metal flow velocity, and forging vibration data, a target flow velocity is determined. Then, based on the metal flow resistance, metal flow velocity, forging vibration data, and lateral force, an axial strain coefficient and a radial strain coefficient are determined. Then, based on the axial strain coefficient and the target flow velocity, a slider control instruction is generated (for controlling the slider to apply pressure to the metal along the axis of the automotive fuel injector forging to form it). Based on the radial strain coefficient, a push rod control instruction is generated (for adjusting the pressure of four push rods to guide the flow of metal in the radial direction. The four push rods are all located in the horizontal direction and are respectively distributed in the positive and negative axis directions of the x-axis and the positive and negative axis directions of the y-axis). Finally, according to the slider control instruction and the push rod control instruction, a forging device and a hydraulic device are respectively controlled to forge the automotive fuel injector forging. This method achieves precise control of metal flow behavior by comprehensively considering metal flow resistance, flow velocity, vibration parameters and lateral force, thereby significantly improving the internal structural uniformity and forming accuracy of the forging. This method regulates the movement of the slider and the push rod through axial and radial strain coefficients, which can effectively control the strain path and deformation coordination, and reduce the probability of forming defects. This method controls the forging process through multi-parameter dynamic optimization, which can reduce unnecessary energy consumption and debugging time, improve production efficiency and system operation stability, and enhance the intelligence and automation level of the forging process. The use of four push rod structures in the horizontal direction, combined with radial strain coefficient control, can effectively control the uniform filling of metal in multiple radial directions, and improve the coaxiality and structural symmetry of the forging.

[0007] In a second aspect, an embodiment of the present application provides a forging device for an automobile fuel injector forging, comprising: an acquisition unit, configured to acquire metal flow resistance, metal flow velocity, forging vibration data, and lateral force; wherein the forging vibration data includes die vibration amplitude and die vibration frequency; the lateral force is used to characterize the intensity of metal radial flow, and the lateral force includes expansion thrust along the x-axis direction and expansion thrust along the y-axis direction; a target flow rate determining unit, configured to determine a target flow rate based on the metal flow resistance, the metal flow rate, and the forging vibration data; a gauge factor determining unit, configured to determine an axial gauge factor and a radial gauge factor based on the metal flow resistance, the metal flow velocity, the forging vibration data, and the lateral force; a control instruction generating unit, configured to generate a slider control instruction based on the axial strain coefficient and the target flow rate, and to generate a push rod control instruction based on the radial strain coefficient; wherein the slider control instruction is configured to control the slider to apply pressure to the metal along the axis of the automobile fuel injector forging to form the forging, and the push rod control instruction is configured to adjust the pressure of four push rods to guide the flow of the metal in the radial direction, wherein the four push rods are all located in a horizontal direction and are respectively distributed in the positive and negative axis directions of the x-axis and the positive and negative axis directions of the y-axis; The forging unit is used to control the forging device and the hydraulic device to forge the automobile fuel injection nozzle forging according to the slider control instruction and the push rod control instruction.

[0008] In a third aspect, an embodiment of the present application provides a forging device for automobile fuel injector forgings, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the method described in any one of the embodiments of the first aspect when executing the computer program.

[0009] It can be understood that the beneficial effects of the second to third aspects mentioned above can be found in the relevant description of the first aspect mentioned above, and will not be repeated here. BRIEF DESCRIPTION OF THE DRAWINGS

[0010] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the following briefly introduces the drawings required for use in the embodiments or descriptions of the prior art. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0011] Figure 1 This is a schematic flow chart of a method for forging an automobile fuel injector forging provided in one embodiment of the present application; Figure 2 It is a structural schematic diagram of the automobile fuel injector forging forging equipment provided in an embodiment of the present application. DETAILED DESCRIPTION

[0012] In the following description, specific details such as specific system structures and techniques are provided for purposes of illustration rather than limitation to facilitate a thorough understanding of the embodiments of the present application. However, it will be apparent to those skilled in the art that the present application may be implemented in other embodiments without these specific details. In other cases, detailed descriptions of well-known systems, devices, circuits, and methods are omitted to avoid obscuring the description of the present application with unnecessary detail.

[0013] It should be understood that when used in the present specification and the appended claims, the term "comprising" indicates the presence of described features, integers, steps, operations, elements and / or components, but does not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components and / or collections thereof.

[0014] It will also be understood that the term "and / or" used in this specification and the appended claims refers to and includes any and all possible combinations of one or more of the associated listed items.

[0015] In related technologies, metal materials often exhibit differences in plasticity, flow resistance, and strain hardening behavior across different batches, heating states, or mold environments. Traditional forging processes generally employ fixed loading paths and preset process parameters, and repeated runs with the same loading paths and motion trajectories in actual production can result in excessive or insufficient loading, leading to insufficient or excessive metal flow. Parts can also experience uneven deformation in localized areas, especially in complex transition zones like the needle valve seat, bevels, and arcs of fuel injectors. Overall forming accuracy is difficult to guarantee, impacting part consistency and subsequent machinability.

[0016] Injector nozzle forgings typically feature complex, small cross-sections and precise contours, such as nozzle guides and tapered connection sections. These areas place extremely high demands on metal filling behavior. However, in traditional forging processes, since the forming process relies solely on single-directional loading control, defects such as insufficient filling and folding of the material within the die cavity are prone to occur. Furthermore, under a fixed loading path, if the metal flow behavior deviates from the preset path, surface and internal defects are easily formed at die corners or flow diversion areas.

[0017] Due to material fluctuations and uncontrollable flow paths, the pressure distribution and metal contact behavior of the same set of molds in different forging cycles vary greatly, which can easily cause premature wear or even cracks in certain areas of the mold due to repeated impacts; product consistency deteriorates, and subsequent inspection and repair processes need to be increased; overall production efficiency decreases, and the pass rate fluctuates greatly, making it difficult to meet large-scale, high-quality forging requirements.

[0018] To address the aforementioned issues, embodiments of the present application provide a method and apparatus for forging automotive fuel injector nozzle forgings. This method first obtains metal flow resistance, metal flow velocity, forging vibration data (die vibration amplitude and die vibration frequency), and lateral force (characterizing the strength of radial metal flow, including expansion thrust along the x-axis and expansion thrust along the y-axis). A target flow velocity is then determined based on the metal flow resistance, metal flow velocity, and forging vibration data. Furthermore, the axial and radial strain coefficients are determined based on the metal flow resistance, metal flow velocity, forging vibration data, and lateral force. Next, based on the axial strain coefficient and target flow velocity, a slider control instruction is generated (for controlling the slider to apply pressure to the metal along the axis of the automotive fuel injector nozzle forging to form it). Furthermore, based on the radial strain coefficient, a push rod control instruction is generated (for adjusting the pressure of four push rods to guide radial metal flow. The four push rods are horizontally located and distributed along the positive and negative x-axis and the positive and negative y-axis directions). Finally, based on the slider and push rod control instructions, the forging device and hydraulic device are controlled to forge the automotive fuel injector nozzle forging. This method achieves precise control of metal flow behavior by comprehensively considering metal flow resistance, flow velocity, vibration parameters and lateral force, thereby significantly improving the internal structural uniformity and forming accuracy of the forging. This method regulates the movement of the slider and the push rod through axial and radial strain coefficients, which can effectively control the strain path and deformation coordination, and reduce the probability of forming defects. This method controls the forging process through multi-parameter dynamic optimization, which can reduce unnecessary energy consumption and debugging time, improve production efficiency and system operation stability, and enhance the intelligence and automation level of the forging process. The use of four push rod structures in the horizontal direction, combined with radial strain coefficient control, can effectively control the uniform filling of metal in multiple radial directions, and improve the coaxiality and structural symmetry of the forging.

[0019] The automobile fuel injector forging method provided in the embodiment of the present application can be applied to automobile fuel injector forging equipment. At this time, the automobile fuel injector forging equipment is the executor of the automobile fuel injector forging method provided in the embodiment of the present application. The embodiment of the present application does not impose any restrictions on the specific type of automobile fuel injector forging equipment.

[0020] For example, Figure 2As shown, the four ejector pins in the figure act horizontally on the die, while the slider acts vertically. The die in the figure is the same automotive fuel injector die. The dashed box in the figure indicates that the die is not part of the hydraulic system. Forging equipment for automotive fuel injector forgings can include a forging device, a hydraulic system, and a control device that communicates with the forging and hydraulic systems. The forging device is a device that applies axial (vertical) pressure to metal via a slider to achieve integral shaping within the automotive fuel injector die cavity. It can be a mechanical press, servo press, or hydraulic forging press. The hydraulic system is a device that directs metal flow in a radial (horizontal) direction to form the die. It can include a fluid storage container (such as a fuel tank), a power element (such as a hydraulic pump), a main control element (such as a relief valve or safety valve), four branch control elements (such as a pressure valve, a flow valve, and a directional valve), and four actuators (such as four hydraulic cylinders (where the ejector pins are located)). The liquid reservoir is connected to the power element via an oil suction pipe, which in turn is connected to the master control element via a pressure oil pipe. The master control element's oil outlet is connected to four branch control elements via distribution lines. Each branch control element is connected to a corresponding actuator via an independent pressure oil pipe. Each actuator is connected to the liquid reservoir via a separate or shared return oil line. The four actuators are evenly distributed horizontally around the mold, mounted on the positive and negative x-axis (+X, -X) and the positive and negative y-axis (+Y, -Y). The power element provides pressurized oil flow to the master control element, which distributes the pressurized oil flow to the four branch control elements. The four branch control elements independently control the opening, closing, and flow rate of their respective oil circuits. Each branch control element adjusts the pressure value of the corresponding actuator, controlling the extension or retraction of the corresponding actuator's ejector rod.

[0021] The overall shape of an automotive fuel injector mold essentially mirrors the shape of the desired fuel injector, consisting of a main cavity and possibly multiple sub-cavities, channels, or internal holes. The mold surface is provided with four holes that mirror the shape of the ejector pins (the shape of the contact surface between the ejector pin and the mold). A ejector pin is placed in each hole, and each ejector pin enters the mold through a channel and comes into contact with the metal inside. The ejector pins are tightly connected to the mold surface through the channels but do not directly clamp the mold. Instead, they use pressure to push the metal along the mold cavity. During the main forging process (where the forging device applies vertical pressure to the metal), the ejector pins remain fixed in place and remain stuck to the mold's inner surface, preventing metal from escaping through the mold holes.

[0022] The control device is a device capable of controlling the forging device and the hydraulic device and performing data processing, and can be a tablet computer, a laptop computer, an ultra-mobile personal computer (UMPC), a netbook, a personal digital assistant (PDA), a desktop computer, a computing device or other processing device connected to a wireless modem, a computer, a laptop computer, a customer premises equipment (CPE) and / or other devices for communicating on a wireless system and a next-generation communication system, for example, a mobile terminal in a 5G network or a mobile terminal in a future evolved public land mobile network (PLMN).

[0023] In order to better understand the forging method of the automobile fuel injector forging provided in the embodiment of the present application, the specific implementation process of the forging method of the automobile fuel injector forging provided in the embodiment of the present application is exemplarily introduced below.

[0024] Figure 1 A schematic flow chart of a forging method for an automobile fuel injector forging provided in an embodiment of the present application is shown. The forging method for an automobile fuel injector forging comprises: S100: Obtain metal flow resistance, metal flow velocity, forging vibration data, and lateral force. The forging vibration data includes die vibration amplitude and die vibration frequency. The lateral force characterizes the strength of radial metal flow and includes expansion thrust along the x-axis and expansion thrust along the y-axis.

[0025] Metal flow resistance refers to the internal resistance to metal flow during the forging process, when the metal undergoes plastic deformation under external forces. This resistance is expressed as stress per unit area. The greater the metal flow resistance, the harder it is to deform.

[0026] For example, a strain gauge or force sensor can be installed on the contact surface between the mold and the metal to monitor in real time the resistance per unit area (i.e., flow stress) experienced by the metal during deformation. The data can be transmitted to the control system through a data acquisition system, and the flow resistance of the metal at a specific time and position can be obtained through calculation.

[0027] It can be understood that metal flow rate refers to the movement speed of metal material when it undergoes plastic flow under the action of forging force. It directly reflects the rate of metal deformation and filling in the mold cavity. It is expressed as the distance the metal flows per unit time, and the unit can be mm / s or m / s.

[0028] For example, a three-dimensional laser Doppler velocimeter (3D LDV) or a three-dimensional laser displacement sensor array can be arranged on the mold surface or in key areas of the metal flow path to collect the flow velocity of the metal in the x, y, and z directions. Based on the flow velocity in the x, y, and z directions, the modulus of the total velocity vector is calculated to be the metal flow velocity. The calculation formula is: ,in, Indicates the metal flow rate, represents the flow velocity in the x-axis direction, represents the flow velocity in the y-axis direction, Indicates the flow velocity in the z-axis direction.

[0029] It can be understood that the forging vibration data describes the periodic vibration characteristics generated by the mold during the forging process, which can include the mold vibration amplitude and mold vibration frequency. The mold vibration amplitude indicates the maximum displacement generated by the mold during the vibration process, and the unit is millimeter (mm); the mold vibration frequency indicates the number of vibrations per unit time, and the unit is Hertz (Hz).

[0030] For example, a high-precision acceleration sensor (such as a piezoelectric accelerometer) or laser vibration sensor can be installed on the mold. The sensor signal is fed into a signal processing module via a data acquisition card (DAQ). This module records the time domain signal (acceleration-time curve) of the mold's acceleration over time during the forging process. The acquired acceleration curve is numerically integrated to obtain a velocity curve, which is then integrated again to obtain a displacement curve. The maximum absolute value of the displacement curve is the mold vibration amplitude.

[0031] The fast Fourier transform (FFT) algorithm can be used to convert the time domain signal into a frequency domain signal, and the frequency component with the largest amplitude in the frequency domain signal is extracted as the main vibration frequency (mold vibration frequency).

[0032] It can be understood that lateral force refers to the force exerted on the die wall along the x- and y-axes when the metal expands radially after being subjected to axial pressure during the forging process. It is used to quantify the metal's radial flow intensity and tendency. The expansion thrust along the x-axis represents the radial force exerted on the metal along the x-axis during the forging process; the expansion thrust along the y-axis represents the radial force exerted on the metal along the y-axis during the forging process.

[0033] For example, four highly sensitive pressure sensors are installed on the die's x- and y-axes. The x-axis pressure sensor measures the expansion thrust along the x-direction, while the y-axis pressure sensor measures the expansion thrust along the y-direction. During the forging process, the metal expands radially under pressure, pushing against the die wall and generating a reaction force. The pressure sensors convert these expansion thrusts in the four directions into electrical signals for recording. The measured values ​​from the x- and y-axis pressure sensors are then averaged or weighted to calculate the expansion thrust along the x- and y-axes.

[0034] Through the simultaneous acquisition and processing of the above four data, comprehensive and accurate input parameters can be provided for subsequent forging process control, laying the foundation for intelligent forging control.

[0035] S200 , determining a target flow rate based on the metal flow resistance, the metal flow rate, and the forging vibration data.

[0036] For example, the metal flow rate is affected by multiple factors. The greater the flow resistance, the harder it is for the metal to flow, and the target flow rate can be reduced accordingly. The greater the vibration amplitude and vibration frequency, the more obvious the softening and lubrication effect on the metal, and the target flow rate can be appropriately increased. The current metal flow rate provides the basis for the flow trend, and the target flow rate can be fine-tuned based on the current metal flow rate. A weighted model or a data-driven model can be used to construct a flow control model, such as ,in, Indicates the target flow rate, Indicates the current metal flow rate, Represents the adjustment coefficient, which can be obtained through experience or data training. Indicates the vibration amplitude of the mold, Indicates the vibration frequency of the mold, Indicates the resistance to metal flow.

[0037] The metal flow resistance, metal flow velocity, and forging vibration data obtained in step S100 can be substituted into the flow control model to calculate the target flow velocity. The target flow velocity serves as a key input parameter for subsequent control strategies, such as slider speed control and deformation rate matching, to facilitate metal flow and mold filling at the optimal speed.

[0038] This step dynamically calculates a target flow rate that makes the forming process more stable, uniform, and defect-free through a comprehensive analysis of the metal's current state (metal flow resistance, metal flow rate) and external excitation (forging vibration data). This provides a quantitative basis for subsequent pressure regulation and path control, and is a key link in realizing intelligent forging.

[0039] In one possible implementation, S200, determining a target flow rate based on metal flow resistance, metal flow rate, and forging vibration data, includes: S210 , determining a rheological resistance control coefficient based on the metal flow resistance, metal flow velocity, and forging vibration data.

[0040] For example, during the hot forging process, the plastic deformation behavior of the metal is affected by the rheological properties of the material itself and the external loading conditions. In order to accurately predict and control the metal deformation behavior, the rheological resistance control coefficient can be dynamically calculated as an important parameter for constructing the forming force model and control strategy. The calculation model of the rheological resistance control coefficient can be constructed by introducing metal flow resistance, metal flow velocity and forging vibration data, that is, ,in, represents the rheological resistance control coefficient, represents the basic rheological control coefficient, Represents the metal flow resistance, Indicates the metal flow rate, Indicates the vibration amplitude of the mold, Indicates the vibration frequency of the mold, and Represents the empirical adjustment coefficient, which is set according to the material type and process conditions.

[0041] By performing a hot compression test or a tensile test on the target material under high temperature conditions, its stress-strain-strain rate relationship curve can be obtained, thereby fitting the constitutive model parameters of the material and extracting the basic rheological coefficient. , represents the plastic resistance response of the material under non-vibration conditions.

[0042] The rheological resistance control coefficient can be calculated based on the acquired metal flow resistance, metal flow rate and forging vibration data and the calculation model of the rheological resistance control coefficient, which reflects the actual response degree of the metal to deformation under the current vibration, flow rate and resistance.

[0043] This step achieves more accurate modeling of metal plastic behavior by introducing metal flow resistance, metal flow velocity and forging vibration data to jointly regulate rheological parameters, providing a dynamic response basis for subsequent intelligent forging control strategies.

[0044] S220: Determine a forming driving force based on the metal flow resistance and the target forming force, wherein the forming driving force is used to represent the net force actually used for metal deformation.

[0045] It can be understood that during the forging process, the total force output by the equipment (i.e., the target forming force) can be decomposed into metal flow resistance (to overcome resistance to flow such as internal friction of the material, mold friction, and interface adhesion) and forming driving force (representing the net effective force actually used by the equipment to drive the material to undergo plastic deformation).

[0046] For example, the total forming load (target forming force) required at the current stage can be determined based on process design, empirical values ​​or finite element simulation results. The force obtained by subtracting the metal flow resistance from the target forming force is the forming driving force, that is, ,in, represents the forming driving force, Indicates the target forming force.

[0047] S230 , determining a target flow rate based on the rheological resistance control coefficient and the molding driving force.

[0048] For example, the target flow rate achieved by the current forming process can be obtained by inversely solving the relationship between the rheological resistance control coefficient, the forming driving force, and the target flow rate. The relationship between the rheological resistance control coefficient, the forming driving force, and the target flow rate is: The target flow rate can be used to generate subsequent control instructions to ensure uniform plastic deformation while taking into account both cavity filling quality and energy efficiency.

[0049] Through the above steps, at each forging time node, the equipment output force is clearly decomposed into the part of overcoming resistance and the part of effective deformation, the material properties and the forming process are organically coupled, and the target flow rate is calculated, thereby achieving precise control and intelligent forging.

[0050] S300, determining the axial strain factor and the radial strain factor based on the metal flow resistance, the metal flow velocity, the forging vibration data and the lateral force.

[0051] For example, the metal flow resistance, metal flow velocity, forging vibration data and lateral force can be transformed into the [0,1] interval through a normalization function to construct a standard input vector to eliminate the dimension effect and facilitate subsequent modeling, such as X=[ 、 、 、 、 、 ],in, Represents the metal flow resistance, Indicates the metal flow rate, Indicates the vibration amplitude of the mold, Indicates the vibration frequency of the mold, represents the expansion thrust along the x-axis, Represents the expansion thrust along the y-axis.

[0052] Data-driven modeling methods such as multivariate regression, neural networks, or support vector machines (SVM) can be used to establish a nonlinear mapping relationship between input parameters and strain coefficients. The axial strain coefficient calculation model is ,in, It represents the axial strain coefficient, which characterizes the unit strain of the metal in the direction of the forging axis (i.e. the direction of the slider), and reflects the degree of axial deformation. The calculation model of the radial strain coefficient is ,in, The radial strain factor reflects the metal's ability to flow and expand in the radial direction (perpendicular to the axial direction) of the mold, and determines the extent of the ejector's supplementary effect. Computational model training can be based on actual production data or finite element simulation data. Model parameters are optimized using an error backpropagation algorithm or least squares method to ensure that the error between the predicted and true values ​​is within an acceptable range.

[0053] Since the various influencing factors in the forging process have different dominant degrees, weight factors can be introduced for coupling adjustment. For example, for metals with high flow resistance, the weight of the vibration frequency in the model can be increased to promote forming; when the lateral force 、 When the strain is greater than the set threshold, the model can strengthen the radial strain weight to avoid metal overflow or insufficient filling of the mold cavity.

[0054] During the forging process, the metal flow resistance and vibration state can be monitored in real time and fed back to the control system through industrial sensors. The system updates the input vector X cyclically based on the latest data and updates the gauge factor in real time. and , thereby achieving dynamic and adaptive deformation control.

[0055] This step provides core parameter support for the generation of subsequent control instructions, which is conducive to the forging achieving the optimal forming state in both the axial and radial directions.

[0056] In one possible implementation, S300 , based on metal flow resistance, metal flow velocity, forging vibration data, and lateral force, determining the axial strain factor and the radial strain factor includes: S310 , based on the current control cycle, obtains time window data before the current control cycle, and predicts the defect probability of the automobile fuel injector forging based on the time window data. The time window data includes a time series of metal flow resistance, a time series of metal flow velocity, and a time series of forging vibration data.

[0057] For example, a fixed-length time window [t0, t] can be set to capture the process dynamics before the current control cycle t. The window length Δt (t-t0) can cover the characteristic change cycle within a typical forging stage (such as pre-pressing, main pressing or vibration-assisted stage) and can be adjusted according to the forging rhythm, such as setting it to 0.5~2 seconds.

[0058] The metal flow resistance data, metal flow velocity data and forging vibration data in the time window can be read from the data storage module in real time, and synchronized with the timestamp to form the metal flow resistance time series, metal flow velocity time series and forging vibration data time series, and the three types of time series are uniformly encapsulated as a time series input matrix. ,in, represents the metal flow resistance time series, represents the metal flow velocity time series, represents the mold vibration amplitude time series, Represents the mold vibration frequency time series.

[0059] Preprocessing and feature extraction of time window data allows calculation of the mean, standard deviation, rate of change, and extreme value locations of the time window data. Fast Fourier transforms (FFTs) can be used to identify periodic anomalies, and time delay embedding or sliding window interpolation can be used to enhance short-term dynamic expression. Time series modeling methods can be used to learn and predict the extracted features. Optional models include LSTM (Long Short-Term Memory) networks, suitable for learning continuous process data with strong temporal dependencies; GRU, 1D-CNN, or Transformer, which balance real-time performance with prediction accuracy; and random forests / support vector machines (SVMs), lightweight solutions for rapid deployment (statistical features must be extracted first). Predictive model training can be based on historical labeled datasets (defective forgings), using defect labels as supervisory signals to construct classification or regression models.

[0060] After the model training is completed, the time series can be input into the matrix X and input into the prediction model. The prediction model can predict the defect probability. If the model has a multi-classification structure, it can also output the defect type (such as insufficient filling, tissue stratification, eccentricity, etc.).

[0061] By building a defect prediction model based on time window data, potential defect risks can be identified before forming is complete, and intelligent forging quality control can be achieved through dynamic adjustment. This mechanism improves the consistency and yield rate of injector forgings and is a key component of building a closed-loop quality management system for intelligent manufacturing.

[0062] Optionally, in step S310, based on the time window data, the defect probability of the automobile fuel injection nozzle forging is predicted, including: S311, performing energy integration on the metal flow resistance time series and the metal flow velocity time series to obtain a total deformation energy.

[0063] For example, the work (i.e., energy) per unit time can be expressed as , continuous sampling in the entire time window, and the total deformation energy is obtained by integrating the power, that is, ,in, represents the total deformation energy.

[0064] The cumulative energy of a time series can be calculated using discrete numerical integration methods. For example, the trapezoidal integration method has the following calculation formula: , the left rectangle method (when the data accuracy is sufficient), the calculation formula is ,in, represents the sampling time interval, Represents the total number of data points (sample points) in the time series.

[0065] By performing point-by-point product integration on the time series of metal flow resistance and flow velocity, the actual deformation energy within the time window can be accurately calculated. This not only quantifies the process energy consumption, but also provides key basic data for forging quality control and energy efficiency optimization.

[0066] S312: Based on the metal flow resistance time series, the resistance mean and resistance standard deviation are calculated, and the ratio between the resistance standard deviation and the resistance mean is calculated to obtain the force fluctuation coefficient.

[0067] For example, the resistance mean represents the average deformation resistance of the metal under controlled loading in the time window. The mean of all metal flow resistances in the metal flow resistance time series can be calculated, which is the resistance mean. ,in, Indicates the mean resistance value.

[0068] The standard deviation of resistance indicates the degree to which the metal flow resistance fluctuates around the mean value, and its calculation formula is: ,in, represents the mean standard deviation.

[0069] The ratio of the standard deviation of the resistance to the mean resistance can be calculated to obtain the force fluctuation coefficient, that is, ,in, It represents the force fluctuation coefficient and is a dimensionless indicator that can be used to measure the smoothness of the forging process.

[0070] S313, based on the metal flow velocity time series, determine whether the difference between adjacent flow velocities is greater than a threshold, count the number of times the difference between adjacent flow velocities is greater than the threshold, and obtain the number of flow velocity mutations.

[0071] For example, a velocity change threshold (e.g., 0.5) can be set to identify abnormal fluctuations. The velocity at adjacent time points is differentially calculated to obtain the difference between the adjacent velocity values. It is then determined whether the difference is greater than the velocity change threshold. If so, it is determined to be a sudden change event. The cumulative number of times that all judgment conditions are met is the number of velocity sudden changes. ,The number of flow velocity mutation changes can be used as an auxiliary indicator for ,abnormality prediction, vibration control judgment, and mold damage risk assessment.

[0072] S314, calculating vibration energy according to the forging vibration data time series.

[0073] For example, under the ideal elastic vibration model, the instantaneous energy density can be expressed as , the instantaneous vibration energy density in the time window can be integrated over time to obtain the vibration energy, that is, ,in, represents vibration energy, represents the equivalent vibration mass (which can be regarded as a constant or obtained through modeling), which can be set =1 is used as a normalized reference to obtain the relative vibration energy index for comparison under different working conditions.

[0074] By statistically analyzing the number of mutations in the metal flow rate time series and quantitatively calculating the vibration energy, we have achieved a quantitative analysis of two important dynamic factors in the forging process, providing a directly usable data basis for process control, defect prediction, and intelligent parameter adjustment.

[0075] S315, constructing a defect vector based on the total deformation energy, the force fluctuation coefficient, the number of flow velocity mutations, and the vibration energy.

[0076] For example, the total deformation energy, force fluctuation coefficient, number of flow velocity mutations, and vibration energy can be normalized (such as min-max normalization or Z-score normalization), and the normalized total deformation energy, force fluctuation coefficient, number of flow velocity mutations, and vibration energy can be spliced ​​into a defect vector .

[0077] S316, based on the defect vector, using a convolutional neural network to predict the defect probability of the automobile fuel injector forging to obtain the defect probability.

[0078] For example, the defect vector can be input into a trained convolutional neural network, which can predict the current defect probability in real time.

[0079] The training process of the convolutional neural network is as follows: process data can be collected from historical forging batches, and four key indicators, namely total deformation energy, force fluctuation coefficient, number of flow velocity mutations, and vibration energy, can be extracted to construct a defect vector. Each set of defect vectors corresponds to a label (sample pair). If the forging has defects (such as insufficient filling or cracks), the label is 1; if the forging has no defects, the label is 0.

[0080] A one-dimensional convolutional neural network (1D-CNN) can be used to model the feature relationship in the defect vector. The structure can include an input layer, where the input can be a four-dimensional feature vector; a convolution layer (Conv1D), which can use multiple 1D convolution kernels to extract local feature combinations (such as 16 convolution kernels with kernel size = 2); an activation function ReLU, which can improve nonlinear modeling capabilities; a pooling layer (MaxPooling1D), which can reduce dimensionality and enhance model stability; a fully connected layer (Dense), which can combine and classify the extracted features; and an output layer, which is a single neuron and can use a Sigmoid activation function to output the defect probability.

[0081] All sample pairs can be divided into a training set (approximately 70%), a validation set (approximately 15%), and a test set (approximately 15%). Binary cross entropy can be used as the loss function, and the Adam optimizer can be used for backpropagation and weight updates. The learning rate can be set to 0.001 and automatically adjusted to accelerate convergence. The network hyperparameters batch size can be set to 32; epochs can be set to 50-100, with early stopping configured based on the validation set loss curve. Dropout layers can be added before fully connected layers to prevent overfitting; and L2 regularization can be applied to the convolution kernels to enhance generalization.

[0082] You can plot the loss function and accuracy curves for the training and validation sets to determine whether overfitting is occurring (increasing validation set loss while decreasing training set loss). You can also save the model parameters that perform best on the validation set (model checkpoint). You can evaluate the trained model on the test set using metrics such as accuracy, precision, recall, and F1 score. If performance is satisfactory, you can deploy the convolutional neural network to the Forging system for real-time prediction.

[0083] By constructing defect vectors that reflect abnormal energy behavior and process disturbances, and combining them with a lightweight convolutional neural network structure, they achieve rapid and accurate prediction of defect probabilities in automotive fuel injector forgings. This approach not only offers high real-time performance and scalability, but can also be integrated as a key quality monitoring module in intelligent forging control systems.

[0084] S320, calculating a real-time filling rate of the mold cavity according to the metal flow rate and the mold cavity pressure.

[0085] For example, in the forging process for automotive fuel injector forgings, the mold cavity fill rate is a key parameter that measures whether the metal fully occupies the mold space and whether there are quality issues such as underfilling and dead corner defects. The real-time fill rate represents the ratio of the volume of the mold filled with metal at the current moment to the total effective volume of the mold. The total effective volume of the mold is a fixed value determined by the forging structure.

[0086] The volume of metal entering the mold cavity per unit time can be calculated based on the metal flow rate. If the initial time is t0, the cumulative inflow volume in the time period [t0, t] is ,in, Represents the volume of metal flowing into the mold cavity, Represents the instantaneous effective flow cross-sectional area of ​​the metal entering the die cavity (which can be approximated as the cross-sectional area of ​​the die cavity entrance or dynamically estimated by pressure feedback), Represents the metal flow rate at the die inlet (measured by a displacement sensor or calculated). In numerical implementation, the volume can be calculated using discrete integration within a sliding time window (such as the trapezoidal integration method).

[0087] Piezoelectric sensors can be placed in key areas of the mold cavity (nozzle seat, flange transition area, etc.) to measure the mold cavity pressure. The mold cavity pressure reflects the degree of obstruction of metal filling. The flow rate and pressure change trend can be combined to determine whether the metal encounters variable cross-section retention or dead corner accumulation. If the metal flow rate is high and the mold cavity pressure remains low, it means that the mold cavity is still in the rapid filling stage; if the metal flow rate decreases and the mold cavity pressure increases significantly, it means that the metal has begun to squeeze the remaining unfilled area and the filling is nearly complete. Therefore, a dynamic correction factor can be set based on the mold cavity pressure to adjust the dynamic correction factor. Perform real-time compensation, i.e. ,in, Indicates the volume of metal flowing into the mold cavity after compensation, represents the dynamic correction factor, ∈(0.95,1.05)The specific function form can be obtained by experimental calibration or machine learning fitting.

[0088] The ratio between the volume of metal flowing into the mold cavity after compensation and the total effective volume of the mold can be calculated to obtain the real-time filling rate of the mold cavity, that is, ,in, Indicates the real-time fill rate, Indicates the total effective volume of the mold.

[0089] By integrating the metal flow rate and mold pressure information, a real-time calculation mechanism for the mold filling rate was established, which can dynamically and accurately evaluate the mold cavity filling progress, providing key technical support for the intelligent forging control and defect warning of injector forgings.

[0090] Optionally, S320, calculating a real-time filling rate of the mold cavity according to the metal flow rate and the mold cavity pressure, includes: S321, determining the pressure filling rate according to the mold cavity pressure.

[0091] For example, n pressure sensors may be arranged in multiple key areas (such as the nozzle seat, flange transition area, etc.) in the mold cavity, and each pressure sensor is responsible for collecting metal pressure data of the area in which it is located in real time.

[0092] For each area where the pressure sensor is located, a pressure reference value (standard saturation pressure) can be preset, that is, the pressure reached in the area under the ideal filling state. The pressure reference value can be determined through historical test data, finite element simulation or empirical formula. The pressure reference values ​​of different areas can be different, reflecting their geometric characteristics and forming difficulty.

[0093] To reflect the contribution of different regions to the overall filling state, weighting factors can be introduced. For example, the nozzle area is difficult to form due to limited flow, so a weight of 1.2 can be set; the flange area is relatively easy to form, so a weight of 0.8 can be set; and other areas can be set to 1.0.

[0094] The calculation formula for pressure filling rate can be ,in, Indicates the pressure filling rate, Indicates the The actual pressure value collected by the pressure sensor in real time, Indicates the The standard saturation pressure of the area where the pressure sensor is located; Indicates the The pressure filling rate can reflect whether the mold cavity is close to the saturated compaction state during the current forming process.

[0095] By integrating local pressure sensing, regional importance weighting, and normalization, an effective indicator reflecting the overall filling quality of the mold cavity—the pressure filling rate—was constructed. This indicator has the advantages of strong real-time performance, clear physical meaning, and easy integration into control systems. It can be widely used in intelligent forging, process diagnosis, and quality prediction.

[0096] S322 , determining a filling progress based on the metal flow rate and the total length of the flow path.

[0097] For example, the mold cavity can be analyzed in advance using a three-dimensional CAD model to determine the longest path segment for the metal to flow from the initial contact point to the final filling of the mold cavity. The length of this longest path segment is the theoretical maximum flow distance required for metal filling, that is, the total length of the flow path.

[0098] In order to quantify the progress of metal filling, the metal advancement length in the mold cavity at the current moment can be calculated according to the metal flow rate, and the ratio of this to the total length of the flow path can be calculated to obtain the filling progress percentage, that is, ,in, Indicates the filling progress, Indicates the total distance the metal front has advanced from the start of forming to the current moment, Indicates the total length of the flow path.

[0099] By combining metal flow velocity and CAD path information, an intuitive and physically clear flow front tracking system was constructed. It can not only quantify the filling progress of forgings in real time, but also provide key decision-making basis for forging process control, helping to achieve intelligent and high-quality metal forming.

[0100] S323, obtaining the current average temperature, and compensating the filling progress according to the current average temperature and the reference temperature to obtain temperature compensation data.

[0101] It's understandable that during hot forging, temperature fluctuations can cause thermal expansion and contraction of the metal, directly impacting the accuracy of the fill path length and formed volume estimates. To more accurately reflect the actual fill level within the die cavity, temperature compensation can be applied to the original fill progress.

[0102] For example, a temperature sensor array may be arranged in a metal flow path or a key area of ​​a mold, and temperature data of multiple points may be collected through the temperature sensor array. The average value of the temperature data of the multiple points may be calculated to obtain the current average temperature.

[0103] A reference temperature (e.g., 1150°C) can be pre-set. The reference temperature can be derived from the ideal forming temperature recommended by the material. If the current average temperature is lower than the reference temperature, the actual volume may be smaller than expected (shrinkage). If the current average temperature is higher than the reference temperature, the filling progress may be inflated due to expansion.

[0104] The temperature compensation coefficient can be determined according to the current average temperature and the reference temperature, and the filling progress can be compensated according to the temperature compensation coefficient, that is, ,in, Indicates the filling progress after temperature compensation (temperature compensation data), Indicates the current average temperature. represents the reference temperature, represents the temperature compensation coefficient, Indicates the coefficient of thermal expansion (unit: 1 / °C), which can be obtained from the thermal properties of the material (for example, steel is about 1.2× ~2.0× ).

[0105] This step significantly improves the accuracy of characterizing the actual forming state, providing a more reliable decision-making support basis for the intelligent forging process.

[0106] S324 , weighted fusion of the pressure filling rate, filling progress and temperature compensation data to obtain a real-time filling rate.

[0107] For example, in order to adapt to the perceptual characteristics of different stages of the forging process, the weight distribution can be dynamically adjusted according to the current filling level to form a phased weighting strategy. As shown in the following table: The current working stage can be determined based on the filling progress, and the corresponding weight can be selected from the above table based on the current working stage. 、 、 , for weight and pressure fill rate, weight and filling progress, The real-time filling rate is obtained by weighted summation of the temperature compensation data.

[0108] By integrating pressure field distribution, path tracking progress and thermal field compensation effects, a dynamic weight-driven real-time filling rate estimation system is constructed, which improves the perception accuracy and stability of the forging process and provides solid data support for the construction of an intelligent forging closed-loop control system.

[0109] S330, constructs a state vector based on the defect probability, the real-time filling rate, the metal flow resistance, the metal flow velocity, the forging vibration data, and the lateral force.

[0110] For example, in order to unify parameters of different dimensions and units, each data can be standardized (such as Z-score normalization or minimum-maximum scaling), and the standardized data can be spliced ​​into an 8-dimensional state vector, that is, .

[0111] S340, based on the state vector, using the strategy network to calculate the axial strain coefficient and the radial strain coefficient.

[0112] It can be understood that the strategy network is a deep neural network structure that is used to directly output control decisions based on the state vector. Its function is to predict the deformation strategy executed by the slider and the push rod based on the current state of the forging system. That is, the axial strain coefficient determines the deformation adjustment strength of the slider in the Z-axis direction; the radial strain coefficient determines the adjustment strength of the push rod in inducing metal flow in the x and y directions.

[0113] For example, the policy network can adopt a multi-layer perceptron (MLP) architecture. For example, the input layer is an 8-dimensional state vector, the hidden layer is 2 to 3 layers, each layer has 32 to 128 neurons, the activation function can use ReLU, and the output layer is 2 continuous output values, namely the axial strain coefficient and the radial strain coefficient, which can be normalized to [0, 1] or [–1, 1] by Sigmoid or Tanh. The output form can be ,in, represents the policy function, Represents the trainable parameters of the policy network. The policy network can be trained using supervised learning, using historical process data (including optimal strain outputs) to train the policy network. Alternatively, it can be trained using reinforcement learning, using the forging process as an environment, constructing a reward function (e.g., reducing defect rate, improving filling uniformity), and performing online or offline learning using policy gradient descent (e.g., PPO, DDPG).

[0114] In each control cycle, the input state vector To the strategy network, the strategy network can output the axial strain coefficient in real time and radial strain coefficient .

[0115] This step realizes data-driven decision-making control of complex forging processes, can respond to changes in process status in real time, can adapt to different molds, materials and working conditions, can access more sensor data, and expand the state dimension.

[0116] Optionally, step S340, calculating the axial strain coefficient and the radial strain coefficient using a strategy network based on the state vector, includes: S341, based on the state vector, uses the strategy network to propagate forward to calculate the axial strain mean and the radial strain mean.

[0117] For example, the state vector can be normalized to the interval [0, 1] to facilitate effective training and inference of the neural network. The normalized state vector can be input into the policy network, which processes the normalized state vector using a linear mapping and a ReLU activation function to obtain a 64-dimensional feature representation. The 64-dimensional feature representation is then normalized to improve training stability and generalization capabilities. The normalized 64-dimensional feature representation is then processed using a Tanh activation to obtain a final feature representation (a 32-dimensional feature representation). The final feature representation is then mapped to three strain means: an axial strain mean and a radial strain mean (x-axis strain mean and y-axis strain mean) through a linear transformation.

[0118] This step maps multi-dimensional state perception information into continuous control quantities (mean of axial and radial strains) through a multi-layer feedforward neural network, realizing the intelligent, multi-axis, and multi-factor control and adjustment capabilities of the forging process.

[0119] S342 , performing motion sampling on the axial strain mean value and the radial strain mean value to obtain the sampled axial strain mean value and radial strain mean value.

[0120] For example, in theory, the mean value of the strain output by the policy network is only an expected value, while the actual strain applied to the forging may fluctuate around the expected value. Therefore, the policy network output can be modeled using a normal distribution model for sampling. Specifically, a two-dimensional Gaussian distribution with a mean of the policy network output and a covariance of a diagonal matrix can be constructed to generate sampling values, that is, ,in, represents the mean axial strain after sampling, represents the mean x-axis strain after sampling, represents the mean y-axis strain after sampling, represents the mean axial strain, represents the mean x-axis strain, represents the mean y-axis strain, represents the covariance matrix.

[0121] In order to control the intensity of the sampling perturbation, a diagonal covariance matrix can be defined , each direction corresponds to a fixed standard deviation , which indicates the allowable degree of fluctuation. The larger the standard deviation, the more the sampled value deviates from the mean, and the stronger the system exploration; on the contrary, it is more stable. Here You can choose 0.1, which means the disturbance is small and suitable for fine-tuning in a relatively stable stage.

[0122] The axial and radial directions can be sampled once respectively. For the axial strain control quantity, Sampling nearby, we get ; For radial strain control, and Sampling nearby, we get 、 The sampling value is the axial and radial deformation strain value that the system expects to impose on the metal in the current state, which can be used as a direct input for the generation of subsequent control instructions.

[0123] This step gives the system a certain degree of random perturbation resistance, preventing the strategy from falling into a local optimum. During the strategy training phase, the sampling mechanism supports diverse exploration, which can improve the comprehensiveness of strategy convergence. The sampling mechanism can simulate the perturbations caused by uncontrollable factors in actual equipment, enhancing the control system's adaptability to real-world operating conditions.

[0124] S343 : Constraint processing is performed on the sampled axial strain mean and the sampled radial strain mean to obtain an axial strain coefficient and a radial strain coefficient. The constraint processing includes a range constraint, a rate of change constraint, and a safety constraint. The radial strain coefficient includes an x-axis strain component and a y-axis strain component.

[0125] For example, the numerical range of the gauge coefficient can be limited to the range that is physically permitted and controllable by the device to avoid illegal or unexecutable control quantities due to sampling offset or strategy errors. For example, the axial strain control value is limited to the range of [0.5, 2.0]; the radial strain control value is limited to the range of [0.4, 1.8]. The range constraint can ensure that the strain amount does not exceed the design limit of the mold, material or hydraulic system, avoiding damage to the equipment or causing abnormal flow. For example, is 1.30, then =clip(1.30, 0.5, 2.0)=1.30.

[0126] In order to prevent the gauge factor from changing suddenly in adjacent control cycles, which may cause severe fluctuations in the system or abnormal metal flow, the variation range of the axial and radial strain values ​​can be limited. For example, the maximum allowable variation can be set. (such as 0.3), if the difference between the new sampling value and the previous cycle value exceeds , the change range of the new sample value is limited to The rate of change constraint can effectively smooth the control behavior, improve the continuity and stability of the forming process, and reduce the frequent fine-tuning impact.

[0127] In order to ensure the reasonable flow direction and distribution characteristics of the metal in the critical stage, specific logical safety rules can also be introduced according to the forming process requirements. For example, when the real-time filling rate >15%, and the radial strain coefficient is lower than 0.6, in order to make the lateral distribution of the metal, the radial strain coefficient can be maintained at no less than the previous cycle or increased to the minimum requirement; if the real-time filling rate If the axial strain coefficient has reached 80%, but is still low (e.g., <1.2), the axial strain coefficient can be appropriately increased to promote continued longitudinal forming of the metal and prevent insufficient intermediate compaction. Safety constraints can be based on experience or experimental rules, ensuring the effectiveness and controllability of the forging process under complex boundary conditions.

[0128] The axial strain coefficient and radial strain coefficient will be used to generate control instructions for the slider and ejector pin to achieve fine adjustment and control of metal flow behavior.

[0129] S400 generates slider control instructions based on the axial strain coefficient and target flow rate, and generates ejector control instructions based on the radial strain coefficient. The slider control instructions control the slider to apply pressure to the metal along the axis of the automotive fuel injector forging to form it. The ejector control instructions adjust the pressure of four ejector pins to guide the metal flow in the radial direction. The four ejector pins are located horizontally, one on the positive and negative x-axis and one on the positive and negative y-axis.

[0130] It can be understood that in the process of forging automobile fuel injector forgings, the slider is the actuator that applies the main pressure along the axis of the forging. Its function is to promote the metal to flow along the predetermined path in the confined space, fill the die cavity and complete the final forming.

[0131] Definitions of the axial and radial directions: Assuming the working surface of an automotive fuel injector mold faces upward, the X-axis extends along the left and right sides of the mold, parallel to the working surface. The Y-axis extends along the front-to-back direction of the mold, also parallel to the working surface. The Z-axis is perpendicular to the horizontal plane formed by the X and Y axes, also perpendicular to the working surface. The radial direction is parallel to the horizontal plane formed by the X and Y axes, and the axial direction is parallel to the Z-axis.

[0132] For example, the control objectives of the slider may include achieving stable deformation of the metal in the axial direction at a target flow rate; ensuring that the axial strain is controlled within the process setting range to avoid over-compaction or metal backflow. Based on the control objectives, the ideal motion parameters of the slider (slider pressing speed, slider displacement, slider pressure) can be reversely calculated. Based on the target flow rate and the metal deformation characteristics, the actual pressing speed of the slider can meet the volume conservation principle of the material and take into account the axial strain requirements of the metal. The slider pressing speed calculation formula is ,in, Indicates the speed at which the slider is pressed down. Indicates the target flow velocity of metal in the axial direction, The flow control model and the flow velocity in the z-axis direction can be adjusted , calculate ,Right now .when When it is a positive value, it indicates that the material is compressed and deformed, and the slider can provide additional displacement; when When it is smaller, the slider speed can be relatively slowed down to prevent overpressure.

[0133] The slider displacement can be calculated based on the slider pressing speed and the set deformation time (which can be given by the process cycle). The slider displacement calculation formula is: ,in, represents the slider displacement, Indicates deformation time. The upper limit of the slide stroke and the safe stop range can be set according to the geometry of the die cavity and the structural characteristics of the forging to avoid overshoot of the metal extrusion die or the slide.

[0134] The slider can provide enough pressure to overcome the flow resistance of the metal and factors such as friction and vibration interference. The slider pressure calculation formula is: ,in Indicates the slider pressure, Indicates the safety factor, which can be set by factors such as mold friction, heat loss, etc. The control system can be based on Set the output pressure of the hydraulic device and provide real-time feedback detection to avoid overload.

[0135] The above calculation results can be summarized to form a set of standardized slider control instructions. For example, {"component": "slider","target_velocity": ,"target_displacement": ,"target_pressure": The slider control instructions can be sent to the servo control system or hydraulic control module to guide the slider to execute a specific motion trajectory and pressure control to ensure that the metal completes the required plastic flow and mold cavity filling in the axial direction.

[0136] During the movement of the slider, the actual speed can be continuously monitored and compared with the Deviation between actual pressure and If a significant deviation is detected, the slider control parameters can be adjusted to dynamically correct the instructions, thereby achieving closed-loop control.

[0137] By combining the axial strain coefficient with the target flow rate, the coordinated scheduling of pressure and speed, the unification of forming rhythm and target flow rate, and the precision control and dynamic compensation of mold cavity filling are achieved, providing a basic guarantee for the high-precision and high-consistency forming of complex small forgings such as automobile fuel injectors.

[0138] It is understandable that in traditional forging processes, both the slider and the ejector pin move along the Z-axis (vertical direction). The slider performs the primary deformation task, while the ejector pin is used to eject the finished product or, when necessary, adjust the flow of metal at the bottom of the die cavity. However, this structure has significant limitations: it cannot effectively induce directional flow of metal in the horizontal direction (X and Y axes). This can easily lead to problems such as uneven metal filling, dead-angle voids, or structural segregation, especially in complex or special-shaped forgings. To this end, this solution transforms the traditional single ejector pin structure into a horizontal ejector pin array and achieves localized, directional metal flow induction through differential control.

[0139] By differentially loading four independent ejector pins distributed across the mold's four quadrants (in the positive and negative directions of the horizontal X-axis and the positive and negative directions of the horizontal Y-axis), radial metal flow along the horizontal X and Y axes can be effectively induced. This three-dimensional coupled control strategy significantly enhances the ability to control complex flow paths, facilitating sufficient filling of mold cavity edges and corners.

[0140] By adjusting the relative position and force of the four ejector pins, the metal flow at the bottom of the die can be actively guided, alleviating dead zones, reducing cold shuts and voids, and improving filling integrity. This is particularly effective for complex forgings with asymmetrical shapes or prone to flow deviation, helping to improve dimensional accuracy and structural uniformity.

[0141] This solution does not alter the slider structure or the Z-axis main pressure system; it simply replaces the original single Z-axis ejector with an array of four independently controllable ejectors in the horizontal direction. This modification is a local upgrade with high structural compatibility, requiring no significant modifications to the mainframe or main drive system, making it highly feasible and cost-effective.

[0142] For example, the four ejector rods are composed of four independent hydraulic pistons distributed in the positive and negative axis directions of the horizontal X axis and the positive and negative axis directions of the horizontal Y axis (+X, -X, +Y, -Y). To achieve local differential adjustment, the target pressure coefficient (pressure factor model) of each ejector rod can be calculated based on the radial strain coefficient and the lateral force. The pressure factor model is , , , ,in, 、 、 、 Represents the target pressure coefficient of the ejector in the +x direction, -x direction, +y direction, and -y direction, respectively. To adjust the gain factor, To prevent a small positive number from dividing by zero, the target pressures of the four ejector pins can be calculated based on the set maximum system output pressure and the target pressure coefficients in the four directions, namely ,in, Indicates the target pressure of the ejector rod, Indicates the maximum output pressure. The target pressures of the four ejectors can be encapsulated as standard control instructions, for example, {"component":"ejector_array","control_mode":"independent_pressure","target_pressures":{"X_pos": ,"X_neg": ,"Y_pos": ,"Y_neg": },"feedback_enabled":true,"response_profile":"adaptive_real_time"}.

[0143] Ejector control commands can be sent to the hydraulic servo system, which controls the movement of the four ejector pins individually, enabling them to dynamically adjust the direction of the metal in the die cavity. During the forging process, the actual ejector pin response can be continuously acquired through pressure sensors and displacement sensors and compared with the set target. If metal flow in a certain direction is insufficient, the feedback system can increase the ejector pin pressure in that direction. If the metal flow rate exceeds expectations or causes localized bulging, the corresponding ejector pin can be appropriately withdrawn to prevent overpressure in the die cavity or uncontrolled deformation of the forging.

[0144] By converting the radial strain coefficient into independent pressure control signals for four ejector pins, precise induction and zonal regulation of radial metal flow is achieved. Compared with traditional single-elevator structures, this strategy not only improves mold cavity filling efficiency but also significantly enhances the system's adaptability to complex flow fields.

[0145] In one possible implementation, in step S400, generating a slider control instruction based on the axial strain coefficient and the target flow rate includes: S410 , calculating the slider velocity according to the axial strain coefficient and the target flow velocity.

[0146] For example, the product of the axial strain coefficient and the target flow rate can be calculated as the slider speed, which is calculated as follows: ,in, Indicates the slider speed, represents the axial strain coefficient.

[0147] This step can achieve dynamic and adaptive slider speed control, avoiding uneven flow or defect accumulation caused by simple uniform speed pressing.

[0148] S420 , determining the slider displacement according to the slider speed, and generating a slider control instruction according to the slider displacement.

[0149] For example, the product of the slider speed and the time step of the current control cycle (the time interval between each round of control instruction updates (e.g., 10ms)) can be calculated as the slider displacement. If the slider displacement is a cumulative value, the slider speed can be integrated and gradually updated.

[0150] The current slider initial position plus the calculated slider displacement can be used to determine the next target position of the slider, or the current slider position can be continuously updated for closed-loop control in continuous control. If position control mode is used, the target position can be directly converted into a standard control command format (slider control instruction). The control system can compare the actual position with the target position and output a regulating signal (voltage, current, or oil pressure) to drive the slider into position. If speed and displacement composite control is used, the slider speed and target position can be converted into slider control instructions. The generated slider control instructions can be transmitted to the actuator through specific interfaces, such as the PLC control bus, servo controller API, and hydraulic proportional valve analog signal (voltage or current).

[0151] This step translates theoretically derived slider displacements into actual physical control commands, completing the closed-loop connection between the strategic network control and the forging equipment. By precisely generating slider control commands, high-quality axial forming of the injector nozzle forging is ensured, providing a stable foundation for subsequent radial flow control and defect suppression.

[0152] In another possible implementation, in step S400, generating a push rod control instruction based on the radial strain coefficient includes: S401 , calculating a target pressure based on the x-axis strain component, the y-axis strain component, and the reference pressure, wherein the target pressure includes target pressures of four ejector pins in the positive and negative axis directions of the horizontal x-axis and the positive and negative axis directions of the horizontal y-axis.

[0153] It can be understood that the x-axis strain component represents the deformation demand of the metal along the positive and negative axis directions of the horizontal x-axis (defined as the east-west direction), reflecting the intensity adjustment of the pressure applied to the east and west direction push rods.

[0154] The y-axis strain component represents the deformation demand of the metal along the positive and negative axis directions of the horizontal y-axis, that is, the horizontal direction perpendicular to the x-axis (defined as the north-south direction), which determines the pressure distribution of the push rods in the south and north directions.

[0155] The base pressure is a preset reference pressure determined by material properties, mold geometry and process goals, and serves as the baseline for adjusting the ejector pressure in all directions.

[0156] For example, the theoretical reference pressure can be calculated based on the material constitutive model, that is, ,in, Indicates the theoretical base pressure, represents the mold constraint coefficient, represents the yield strength of the material, Indicates the thickness compression ratio of the forging. The yield strength of the material can be determined based on the yield properties of the material at high temperatures. For example, the yield strength of 42CrMo steel at 1150°C is about 80 MPa. The initial thickness of the forging can be calculated. and final thickness The ratio of compression to force reflects the degree of plastic deformation. For example, the typical compression ratio of an automotive fuel injector is about 2.5. Due to the constraint effect of the die structure, the die constraint coefficient ranges from 1.2 to 1.5 in closed die forging. It is used to correct the theoretical pressure and reflect factors such as friction and die rigidity.

[0157] The theoretical reference pressure can be constrained by the ratio of the maximum thrust of the push rod to the area of ​​a single push rod, that is, the theoretical reference pressure can be less than or equal to the ratio of the maximum thrust of the push rod to the area of ​​a single push rod, which can prevent mechanical damage caused by exceeding the maximum thrust of the equipment. In order to ensure uniform flow of metal in the radial direction, the pressure difference of the push rod in the east-west direction and the north-south direction can be limited to not exceed a certain proportion of the yield strength of the material, such as the rod pressure difference is less than or equal to 0.3 The reference pressure can be obtained by constraining the maximum thrust of the ejector pin and the area of ​​a single ejector pin, as well as limiting the pressure difference, on the theoretical reference pressure.

[0158] The baseline pressure achieved in this way not only meets the requirements for ejector pressure plastic deformation, but also ensures equipment safety and forging quality. This baseline pressure serves as the fundamental parameter for subsequent pressure regulation and is key to achieving high-quality nozzle forging control.

[0159] The target pressures in four directions can be calculated based on the reference pressure and the x-axis strain component and the y-axis strain component by using the linear interpolation and symmetric mapping method. The target pressures in four directions can be calculated based on the linear interpolation structure. , this equation satisfies the fact that the total thrust remains unchanged, and the deformation direction can be controlled by adjusting only the distribution ratio, where Indicates the target pressure of the east-direction push rod, Indicates the target pressure of the west-direction push rod, Target pressure of the south-facing ram, The target pressure of the top rod in the north direction. Taking the east-west direction as an example, , ,in, represents the x-axis strain component, when = 0, symmetrical loading, =1.5 , =0.5 ;when =1, balanced loading, = = ;when =2, antisymmetric loading, =0.5 , =1.5 The target pressure calculation method in the east-west direction represents a symmetric-bias-antisymmetric adjustment method, which can achieve continuous control of the flow direction. Similarly, the calculation method in the north-south direction is , ,in, represents the y-axis strain component.

[0160] Dynamic adjustment of the ejector pin pressure through a linear function can support precise control of radial flow in the intelligent forging system, effectively improving the quality of forgings and the stability of the forming process.

[0161] S402 , obtaining actual pressures in four directions, and calculating the error between the actual pressure and the target pressure in each direction.

[0162] For example, pressure sensors located at four key locations along the positive and negative axes of the horizontal X-axis and the positive and negative axes of the horizontal Y-axis can be used to collect the actual pressure values ​​in the four directions at the current moment. The pressure errors (error values) between the target pressures in the four directions and their corresponding actual pressures are calculated.

[0163] S403: Generate a push rod control instruction in each direction based on the error value in each direction.

[0164] For example, if the error is positive, it means that the actual pressure is lower than the target pressure, indicating that the metal support in this direction is insufficient, and the push rod can be pushed forward to increase the forming force in this area; if the error is negative, it means that the actual pressure is higher than the target pressure, indicating that the force in this area is too large, and the push rod can be slowed down, stopped or retracted to avoid excessive local compaction or folding; if the error is zero or within the tolerance range, it means that the actual pressure is close to the target pressure and can maintain the current state without adjustment.

[0165] The thrust output of the ejector is controlled by hydraulic pressure. The error value can be mapped to the hydraulic valve's adjustment value (e.g., boost, hold, or reduce pressure). A linear proportional gain can be set so that the ejector thrust varies linearly with the error. The error value can also be mapped to a reference value for the thrust speed to control the ejector's thrust rhythm. Larger errors correspond to higher thrust speeds, and the speed is slowed down as the target is approached to prevent overshoot.

[0166] For each ejector direction, the resulting ejector control command can include a target hydraulic pressure or thrust value, ejector propulsion speed, control duration, or a threshold judgment condition (such as stopping upon reaching the target pressure). This command can be issued in real time to the corresponding ejector control valve group via a control system (such as a PLC or hydraulic servo system), achieving precise pressure regulation and propulsion control.

[0167] Through the above steps, the direction of metal flow can be adjusted, the pressure distribution differences in the mold cavity can be compensated, and a more uniform forging process can be achieved. It has feedback closed-loop control capabilities and is a key link in the intelligent control of precision forging.

[0168] S500, according to the slider control instruction and the ejector control instruction, respectively controlling the forging device and the hydraulic device to forge the automobile fuel injection nozzle forging.

[0169] For example, the slider control instruction can be transmitted to the forging device control unit, and the forging device control unit can set the initial position of the slider in the forging device according to the slider control instruction to ensure alignment with the metal billet, start the downward pressing action according to the slider downward speed in the slider control instruction, and automatically decelerate when the displacement reaches the specified value (slider displacement). The downward pressing pressure can be dynamically adjusted to match the slider pressure in the slider control instruction, and the hydraulic valve group in the forging device can be adjusted through pressure sensor feedback. The slider load-stroke curve can be monitored to achieve closed-loop control of the entire forming process.

[0170] The ejector control command can be transmitted to the hydraulic device control unit. The hydraulic device control unit can set the initial extension position according to the ejector control command to ensure that the four ejectors contact the metal blank synchronously. According to the target pressure of the four ejectors, the four hydraulic cylinders in the hydraulic device are driven to output different pressures respectively. During the forming process, the ejector position and force can be adjusted in real time according to the metal flow state to induce the metal to expand in an orderly manner in the radial direction of the die cavity. Through the feedback of displacement and load sensors, differential coordination between the ejectors can be achieved to avoid metal accumulation or deviation.

[0171] The control of the slider and ejector pin does not operate in isolation, but forms a linkage mechanism through system coordination. The control system continuously monitors the metal flow rate, mold stress state and displacement progress. If local slow flow, insufficient filling or overloading is detected, the ejector pin pressure distribution can be adjusted, or the slider stroke can be fine-tuned. At different stages (such as pre-pressure, main pressure, and holding pressure), the control strategy can be switched to adapt to the metal strain behavior and flow characteristics. Through this real-time coordinated control of the slider and ejector pin, highly adaptive three-dimensional space forming pressure field scheduling can be achieved.

[0172] After forming is completed, the slider stops pressurizing and quickly returns to its original position, and the ejector maintains a micro-pressure state to keep the forging in a stable position.

[0173] This step achieves efficient collaboration between the forging and hydraulic systems by using slider control instructions for primary forming control and ejector control instructions for flow guidance and compensation. This significantly improves the cavity filling quality and forming accuracy of complex forgings while maintaining a simple equipment structure, demonstrating its potential for industrial application and adaptability to intelligent manufacturing.

[0174] In one possible implementation, the method for forging an automobile fuel injector forging further includes: S10, constructs a reward function based on the defect probability, real-time filling rate and forging vibration data.

[0175] For example, the reward function can be composed of positive rewards and negative penalties, and can include filling completeness and forming efficiency (positive rewards), defect risk suppression (negative penalties), and vibration stability (negative penalties). The expression of the reward function is ,in represents the reward function, 、 、 Represents the weight coefficient, which can be adjusted through experience or dynamically adapted during training; the weight can be set according to the task preferences at different stages, such as emphasizing filling efficiency in the early stage and focusing more on defects and stability in the middle and late stages.

[0176] This item can be used to encourage the control strategy to fill the cavity as quickly and fully as possible. When the real-time fill rate approaches 1 (i.e., the filling is close to complete), this item tends to the maximum value. , indicating an excellent process; conversely, if the filling is severely insufficient or overfilled, the reward decays rapidly, encouraging the system to quickly return to the target filling rate.

[0177] This term can indirectly optimize the metal flow path and load configuration by suppressing the predicted defect probability. As the defect probability increases, this term decreases rapidly, effectively penalizing potential unreasonable operations or strain distributions.

[0178] This item can be used to suppress severe mechanical vibration or system instability during the forging process. Represents the vibration amplitude or frequency energy within a time window. This term is squared and strongly penalizes severe vibrations.

[0179] This reward function comprehensively considers the real-time fill rate (positive feedback), defect probability (negative feedback), and vibration stability (negative feedback), embodying the multi-objective optimization concept in forging control. By continuously adjusting the control strategy to maximize this reward function, the reinforcement learning algorithm can gradually learn the optimal forming path and force control strategy, thereby improving the quality and consistency of the injector nozzle forgings.

[0180] S20, determining a temporal difference residual according to the reward function and the state vector.

[0181] For example, after the control action is executed, the state vector after the action is executed can be collected. , the state vector is the current state vector , the current state vector and the state vector after the action is executed can be input into the Critic network, and the Critic network can evaluate the value of the current and the state after the action is executed ( 、 ). The Critic network is a value function estimator used to judge the long-term performance value of a system in a certain state.

[0182] The time difference residual (TD error) is used to measure the gap between the current strategy behavior and the expected optimal behavior. Its calculation formula is ,in, Represents the discount factor (e.g., set to 0.95), which is used to control the importance of future returns. According to the TD error calculation formula, the time series difference residual can be calculated.

[0183] S30: updating the policy network parameters based on the time series difference residual, wherein the updated policy network parameters are used to calculate the axial strain coefficient and the radial strain coefficient of the next control cycle.

[0184] It can be understood that the policy network takes the current state vector as input and outputs the distribution probability of an action, that is, ,in, represents the parameters of the current policy network, Indicates the axial strain coefficient and radial strain coefficient output in the current control cycle.

[0185] For example, the time series difference residual can be and strategic networks , calculate the policy gradient, which is used to guide the adjustment direction of the parameters, that is, If the action obtains a high positive feedback (TD error is positive), the probability of the action can be increased; if the feedback is poor (TD error is negative), the probability of the action can be reduced.

[0186] The policy network parameters can be updated based on the policy gradient and a pre-set learning rate (such as 0.0003) ,Right now ,in, Represents the learning rate of the policy network, which is used to control the update amplitude.

[0187] New policy network parameters It will be used in the next control cycle to forward propagate the current state based on the updated strategy network and output better axial strain coefficient and radial strain coefficient to achieve more intelligent and adaptive control of the forging process.

[0188] Through the policy network update mechanism based on reward feedback and state valuation, the system can continuously learn from the operation results, iteratively improve the intelligence of the control strategy and the forming stability, thereby achieving high-quality, low-defect intelligent forging.

[0189] It should be understood that the size of the serial numbers of the steps in the above embodiments does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of this application.

[0190] Corresponding to the forging method of automobile fuel injector forgings described in the above embodiment, the embodiment of the present application also provides a forging device for automobile fuel injector forgings, and each unit of the device can implement each step of the forging method of automobile fuel injector forgings.

[0191] The device includes: The acquisition unit is used to obtain metal flow resistance, metal flow velocity, forging vibration data, and lateral force. The forging vibration data includes die vibration amplitude and die vibration frequency. The lateral force is used to characterize the intensity of radial metal flow and includes expansion thrust along the x-axis and expansion thrust along the y-axis.

[0192] The target flow rate determination unit is used to determine the target flow rate based on the metal flow resistance, the metal flow rate and the forging vibration data.

[0193] The gauge factor determination unit is used to determine the axial gauge factor and the radial gauge factor based on the metal flow resistance, the metal flow velocity, the forging vibration data and the lateral force.

[0194] The control instruction generation unit is used to generate slider control instructions based on the axial strain coefficient and the target flow rate, and to generate ejector control instructions based on the radial strain coefficient. The slider control instructions are used to control the slider to apply pressure to the metal along the axis of the automotive fuel injector forging to form it. The ejector control instructions are used to adjust the pressure of four ejector pins to guide the metal flow in the radial direction. The four ejector pins are all located horizontally, distributed in the positive and negative directions of the x-axis and the positive and negative directions of the y-axis.

[0195] The forging unit is used to control the forging device and the hydraulic device respectively to forge the automobile fuel injection nozzle forging according to the slider control instructions and the push rod control instructions.

[0196] It should be noted that the information interaction, execution process, etc. between the above-mentioned units are based on the same concept as the method embodiment of this application. Their specific functions and technical effects can be found in the method embodiment part and will not be repeated here.

[0197] The embodiment of the present application also provides a forging device for automobile fuel injector forgings, Figure 2This is a schematic diagram of the structure of a forging device for automobile fuel injection nozzles provided in one embodiment of the present application. The forging device for automobile fuel injection nozzles includes a forging device, a hydraulic device, and a control device that is in communication with the forging device and the hydraulic device. Figure 2 As shown, the control device 6 of the automobile fuel injection nozzle forging equipment of this embodiment includes: at least one processor 60 ( Figure 2 Only one is shown), at least one memory 61 ( Figure 2 Only one is shown in the figure) and a computer program 62 stored in the at least one memory 61 and executable on the at least one processor 60. When the processor 60 executes the computer program 62, the automobile fuel injector forging equipment implements the steps of any of the above-mentioned automobile fuel injector forging method embodiments, or implements the functions of each unit in the above-mentioned device embodiments.

[0198] For example, the computer program 62 may be divided into one or more modules / units, which are stored in the memory 61 and executed by the processor 60 to implement the present application. The one or more modules / units may be a series of computer program instruction segments capable of performing specific functions, and the instruction segments are used to describe the execution process of the computer program 62 in the control device 6 of the automobile fuel injector forging equipment.

[0199] The control device 6 of the automobile fuel injector forging equipment can be a computing device such as a desktop computer, a notebook, a palmtop computer, a cloud server, etc. The automobile fuel injector forging equipment can include, but is not limited to, a processor 60 and a memory 61. Those skilled in the art will understand that Figure 2 It is only an example of the forging equipment for automobile fuel injector forgings and does not constitute a limitation on the forging equipment for automobile fuel injector forgings. It may include more or fewer components than shown in the figure, or a combination of certain components, or different components. For example, it may also include input and output devices, network access devices, buses, etc.

[0200] The processor 60 may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or any conventional processor.

[0201] In some embodiments, the memory 61 may be an internal storage unit of the control device 6 of the automotive fuel injector forging equipment, such as a hard drive or memory of the automotive fuel injector forging equipment. In other embodiments, the memory 61 may also be an external storage device of the automotive fuel injector forging equipment, such as a plug-in hard drive, a Smart Media Card (SMC), a Secure Digital (SD) card, a flash memory card, etc. Furthermore, the memory 61 may include both an internal storage unit of the automotive fuel injector forging equipment and an external storage device. The memory 61 is used to store an operating system, application programs, a boot loader, data, and other programs, such as the program code of the computer program. The memory 61 may also be used to temporarily store data that has been output or is about to be output.

[0202] In the above embodiments, the description of each embodiment has its own focus. For parts that are not described or recorded in detail in a certain embodiment, reference can be made to the relevant description of other embodiments.

[0203] The above-described embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. These modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the various embodiments of the present application, and should all be included in the scope of protection of the present application.

Claims

1. A forging method for an automobile fuel injection nozzle forging, characterized in that: include: Obtaining metal flow resistance, metal flow velocity, forging vibration data, and lateral force; wherein the forging vibration data includes die vibration amplitude and die vibration frequency; the lateral force is used to characterize the intensity of metal radial flow; the lateral force includes expansion thrust along the x-axis direction and expansion thrust along the y-axis direction; determining a target flow rate based on the metal flow resistance, the metal flow rate, and the forging vibration data; determining an axial strain factor and a radial strain factor based on the metal flow resistance, the metal flow velocity, the forging vibration data, and the lateral force; Based on the axial strain coefficient and the target flow rate, a slider control instruction is generated, and based on the radial strain coefficient, a push rod control instruction is generated; wherein the slider control instruction is used to control the slider to apply pressure to the metal along the axis of the automobile fuel injector forging to form the forging, and the push rod control instruction is used to adjust the pressure of four push rods to guide the flow of the metal in the radial direction, and the four push rods are all located in the horizontal direction and are respectively distributed in the positive and negative axis directions of the x-axis and the positive and negative axis directions of the y-axis; According to the slider control instruction and the push rod control instruction, the forging device and the hydraulic device are respectively controlled to forge the automobile fuel injection nozzle forging.

2. The method for forging an automobile fuel injection nozzle forging according to claim 1, wherein: The determining of the target flow rate based on the metal flow resistance, the metal flow rate and the forging vibration data comprises: determining a rheological resistance control coefficient based on the metal flow resistance, the metal flow velocity, and the forging vibration data; Determining a forming driving force based on the metal flow resistance and the target forming force; wherein the forming driving force is used to represent the net force actually used for metal deformation; The target flow rate is determined based on the rheological resistance control coefficient and the molding driving force.

3. The method for forging an automobile fuel injection nozzle forging according to claim 1, wherein: The determining of the axial strain coefficient and the radial strain coefficient based on the metal flow resistance, the metal flow velocity, the forging vibration data, and the lateral force comprises: According to a current control cycle, obtaining time window data before the current control cycle, and predicting the defect probability of the automobile fuel injector forging based on the time window data; wherein the time window data includes a metal flow resistance time series, a metal flow velocity time series, and a forging vibration data time series; Calculating a real-time filling rate of the mold cavity based on the metal flow rate and the mold cavity pressure; constructing a state vector according to the defect probability, the real-time filling rate, the metal flow resistance, the metal flow velocity, the forging vibration data, and the lateral force; The axial strain coefficient and the radial strain coefficient are calculated based on the state vector using a strategy network.

4. The method for forging an automobile fuel injection nozzle forging according to claim 3, wherein: The predicting of the defect probability of the automobile fuel injector forging based on the time window data includes: Performing energy integration on the metal flow resistance time series and the metal flow velocity time series to obtain total deformation energy; Based on the metal flow resistance time series, calculating the resistance mean and resistance standard deviation, and calculating the ratio between the resistance standard deviation and the resistance mean to obtain a force fluctuation coefficient; Based on the metal flow velocity time series, determine whether the difference between adjacent flow velocities is greater than a threshold, count the number of times the difference between adjacent flow velocities is greater than the threshold, and obtain the number of flow velocity mutations; Calculating vibration energy according to the forging vibration data time series; constructing a defect vector according to the total deformation energy, the force fluctuation coefficient, the number of flow velocity mutations, and the vibration energy; Based on the defect vector, a convolutional neural network is used to predict the defect probability of the automobile fuel injector forging to obtain the defect probability.

5. The method for forging an automobile fuel injection nozzle forging according to claim 3, wherein: Calculating the real-time filling rate of the mold cavity according to the metal flow rate and the mold cavity pressure includes: determining a pressure filling rate according to the mold cavity pressure; determining a filling progress based on the metal flow rate and the total length of the flow path; Acquiring a current average temperature, and compensating the filling progress according to the current average temperature and a reference temperature to obtain temperature compensation data; The pressure filling rate, the filling progress and the temperature compensation data are weighted and fused to obtain the real-time filling rate.

6. The method for forging an automobile fuel injection nozzle forging according to claim 3, wherein: The calculating the axial strain coefficient and the radial strain coefficient by using a strategy network based on the state vector includes: Based on the state vector, using the strategy network to forward propagate and calculate the axial strain mean and the radial strain mean; Performing motion sampling on the axial strain mean value and the radial strain mean value to obtain a sampled axial strain mean value and a sampled radial strain mean value; The sampled axial strain mean and the sampled radial strain mean are subjected to constraint processing to obtain the axial strain coefficient and the radial strain coefficient; wherein the constraint processing includes a range constraint, a rate of change constraint, and a safety constraint, and the radial strain coefficient includes an x-axis strain component and a y-axis strain component.

7. The method for forging an automobile fuel injection nozzle forging according to claim 1, wherein: The generating of a slider control instruction based on the axial strain coefficient and the target flow rate includes: Calculating a slider velocity according to the axial strain coefficient and the target flow velocity; The slider displacement is determined according to the slider speed, and the slider control instruction is generated according to the slider displacement.

8. The method for forging an automobile fuel injection nozzle forging according to claim 6, wherein: The generating of a push rod control instruction based on the radial strain coefficient includes: Calculating a target pressure based on the x-axis strain component, the y-axis strain component, and a reference pressure; wherein the target pressure includes target pressures of four ejector pins in the positive and negative axis directions of the horizontal x-axis and the positive and negative axis directions of the horizontal y-axis; Get the actual pressure in four directions and calculate the error between the actual pressure and the target pressure in each direction; Based on the error value in each direction, a push rod control instruction in each direction is generated.

9. The method for forging an automobile fuel injection nozzle forging according to claim 3, wherein: The method further comprises: constructing a reward function based on the defect probability, the real-time filling rate, and the forging vibration data; Determining a temporal difference residual based on the reward function and the state vector; Based on the time difference residual, the strategy network parameters are updated; wherein the updated strategy network parameters are used to calculate the axial strain coefficient and the radial strain coefficient of the next control cycle.

10. A forging device for automobile fuel injector forgings, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the computer program, the method according to any one of claims 1 to 9 is implemented.

Images