A Real-time Intelligent Control Method and System for Metal Flow during Forging

Abstract

The present invention belongs to the technical field of forging control, and specifically discloses a real-time intelligent control method and system for metal flow during forging, which includes the digital twin system establishment stage: obtaining dynamic signals and static process parameters during hot forging production, determining the corresponding forming physical data, and then establishing a digital twin system. By means of the digital twin system, the temperature field in the forging is reconstructed, the plastic flow of metal in the forging die is simulated, and forging defects are predicted; the real-time control stage of metal flow: dynamically obtaining signals and static process parameters in real time, and inputting them into the digital twin system to predict the filling rate of each cavity of the current die, the final overall filling rate, and forging defects; taking the small deviation of the filling rate of each cavity of the current die, the high final overall filling rate, and the low forging defects as the optimization goal, and using the genetic algorithm to obtain the local temperature field and slider speed control strategy. The present invention realizes the real-time control of metal flow during the forging process and improves the forming accuracy and quality of parts.

Description

Technical Field

[0001] The present invention belongs to the technical field of forging control, and more specifically, relates to a method and system for real-time intelligent control of metal flow direction during forging. Background Art

[0002] The hot forging process of metals is an important forming process for producing high-quality key structural parts, with advantages such as high forming efficiency, high material utilization rate, and high part quality. However, during the forging process of parts with complex structures, the resistance of metal flow at different positions in the cavity is different, which increases the difficulty of controlling the metal flow direction during forging and results in difficulty in filling the forging cavity at local positions. Therefore, in order to form a complete forging geometry, it is necessary to apply a great forming force at the final stage of forging to squeeze the material into the difficult-to-fill cavity. In this process, the great stress will reduce the die life or even cause die cracking, and it is also prone to uneven flow, resulting in material folding. Therefore, it is necessary to develop a method for real-time control of metal flow direction during forging to achieve a better forging forming effect.

[0003] The rheological behavior of metal materials in thermoplastic forming is greatly affected by the forming temperature and strain rate. Generally, the fluidity of metal materials increases with the increase of temperature and the decrease of strain rate, and the effects of strain rate and temperature on the rheological behavior of materials have a coupling effect. Therefore, by means of local temperature control, increasing the temperature of the position where the material flow is difficult is expected to increase the fluidity of the material in the difficult-to-deform area; at the same time, appropriately reducing the temperature of the cavity that is easy to fill can promote the simultaneous filling of each cavity. Considering the coupling effect of strain rate and temperature, while performing local temperature control, further real-time feedback decision is made on the overall strain rate during the forging of the forging.

[0004] Digital twin is driven by multi-dimensional virtual simulation and fusion data, and through the virtual-real closed-loop interaction, it realizes monitoring, simulation, prediction, optimization, etc., and has been successfully applied in scenarios such as the reliability testing of mechanical products, the intelligent monitoring of production workshops, the assembly technology of complex products, and logistics distribution, achieving online monitoring, simulation, and decision-making of physical entities, with real-time and fidelity.

[0005] Patent CN114372725A proposes a forming monitoring system and method for an additive manufacturing system based on digital twin, realizing the full life cycle management of comprehensively monitoring additive formed parts.

[0006] Patent CN114311826A proposes a fiber metal laminate hydroforming system based on digital twin technology, realizing the precise control of the working parameters of the fiber metal laminate hydroforming equipment to ensure the forming quality and effectively reduce the forming difficulty.

[0007] Patent CN112427624A proposes a digital-twin-based casting and forging dual-control forming system and parameter optimization method. In the digital-twin system, a virtual model of the physical entity is established and dynamic simulation of the casting and forging dual-control forming process is carried out. At the same time, intelligent optimization adjustment is carried out on the key process parameters of the casting and forging dual-control forming process to improve the closed-loop optimization management of the production process. However, the method proposed by this invention can only monitor and optimize the production and manufacturing process of a batch of parts, and cannot carry out real-time regulation on the production of a single part.

[0008] The above method based on the digital-twin production system can realize intelligent optimization and decision-making of the production process, but cannot realize real-time control of the metal flow direction during the forging process. Summary of the Invention

[0009] In view of the above defects or improvement requirements of the prior art, the present invention provides a method and system for real-time intelligent regulation of metal flow direction during forging. The purpose is to realize real-time control of the metal flow direction, ensure uniform filling of the forging cavity, and prevent uneven flow during the forging process.

[0010] To achieve the above object, according to the first aspect of the present invention, a method for real-time intelligent regulation of metal flow direction during forging is proposed, including a digital-twin system establishment stage and a metal flow direction real-time regulation stage, wherein:

[0011] Digital-twin system establishment stage:

[0012] Obtain dynamic signals and static process parameters in the hot forging production process, and determine the corresponding forming physical data to form a data set;

[0013] Based on the data set, establish a digital-twin system, specifically by reconstructing the temperature field in the forging in the digital-twin system, and then simulating the plastic flow of the metal in the forging die to predict defects in the forging process;

[0014] Metal flow direction real-time regulation stage:

[0015] Real-time obtain dynamic signals and static process parameters in the hot forging production process, and input them into the digital-twin system to predict the filling rate of each cavity of the current die, the final overall filling rate, and forging defects;

[0016] Taking the small deviation of the filling rate of each cavity of the current die, the high final overall filling rate, and the low forging defects as the optimization objective, and taking the local temperature field and the slider speed as the optimization variables, use the genetic algorithm to obtain the control strategy, and adjust the local temperature field and the slider speed of the hot forging production in real time according to the control strategy to realize real-time regulation of the metal flow direction during the forging production process.

[0017] As a further preferred option, based on the data set, establish a digital-twin system, specifically:

[0018] In the digital twin system, there are a temperature field reconstruction model, a metal plastic flow prediction model, and a forging defect prediction model, which are respectively used to reconstruct the temperature field in the forging, simulate the plastic flow of metal in the forging die, and predict defects in the forging process; each model is constructed through a data set to obtain a digital twin system used in the real-time regulation stage of metal flow direction.

[0019] As a further preference, the temperature field reconstruction model is constructed as follows:

[0020] Perform finite element simulation on the forging process with different temperature gradients to obtain the discrete point temperatures at the temperature measurement positions in the controllable local temperature control forging die and the overall temperature field distribution.

[0021] Taking the discrete point temperatures and coordinates as inputs and the temperature field distribution as the output, train the temperature field reconstruction model constructed by the BP neural network to obtain a trained temperature field reconstruction model.

[0022] As a further preference, the metal plastic flow prediction model is constructed as follows:

[0023] Conduct thermal simulation experiments with different deformation temperatures and strain rates, establish a stress-strain constitutive model based on the experimental results, and then based on the constitutive model, simulate the hot forging conditions with different overall strain rates and local temperature distributions through finite element method to obtain the metal plastic flow field.

[0024] Taking the overall strain rate and local temperature distribution as inputs and the metal plastic flow field as the output, train the metal plastic flow prediction model constructed by the long short-term memory artificial neural network to obtain a trained metal plastic flow prediction model.

[0025] As a further preference, the forging defect prediction model is constructed as follows:

[0026] Establish a hot processing diagram through thermal simulation experimental data, and then based on this hot processing diagram, obtain the power dissipation rate and instability coefficient of the hot processing output through the metal plastic flow field and the temperature field, so as to judge whether there are forging defects.

[0027] As a further preference, in the real-time regulation stage of metal flow direction, predict the filling rate of each cavity of the current die, the final overall filling rate, and forging defects through the digital twin system, specifically:

[0028] Input dynamic signals and static process parameters into the temperature field reconstruction model to obtain the real-time reconstructed temperature field distribution; then input the temperature field distribution into the metal plastic flow prediction model to simulate the metal plastic flow field, simulate the working state of the equipment, and then predict the filling rate of each cavity and the final overall filling rate of the current die; input the metal plastic flow field and the temperature field into the forging defect prediction model to predict the possible defects during the forging process.

[0029] As a further preference, the dynamic signals include ultrasonic images, local die temperature, forming force, and displacement; the static process parameters include equipment model, blank composition, and forging shape.

[0030] As a further preference, the local temperature field of hot forging production is adjusted by a locally temperature-controlled forging die; the locally temperature-controlled forging die includes a heating system with multiple electric heating wires and a temperature measurement system with multiple thermocouples; the electric heating wires and the thermocouples are arranged near the die cavity to achieve local temperature control and temperature field monitoring.

[0031] According to the second aspect of the present invention, there is provided a real-time intelligent control system for metal flow direction during forging, which includes a processor for executing the above-mentioned real-time intelligent control method for metal flow direction during forging.

[0032] According to the third aspect of the present invention, there is provided a computer-readable storage medium, on which a computer program is stored, characterized in that when the computer program is executed by a processor, the above-mentioned real-time intelligent control method for metal flow direction during forging is realized.

[0033] Generally speaking, compared with the prior art through the above technical solutions conceived by the present invention, the following technical advantages are mainly possessed:

[0034] 1. By collecting dynamic information and static process parameters during the forging process, the present invention constructs a digital twin system for controlling the metal flow direction during forging, realizes the real-time monitoring of the cavity filling situation and the prediction of material flow during the forging process, and makes real-time decisions and controls on the local temperature and the overall strain rate, so as to adjust the local resistance in the metal flow process, realize the intelligent real-time control of the metal flow direction during the forging process, achieve the monitoring and management of the entire production cycle, solve the problems of difficult control of material flow and poor part forming accuracy faced in the forging process of complex parts, and greatly improve the production efficiency and part quality.

[0035] 2. The present invention obtains forming physical data through thermal simulation experiments, finite element simulations, and hot forging experiments, incorporates the multi-source data collected during the hot forging process into the digital twin system, establishes a quantitative model for the cavity filling situation, and a metal flow prediction model related to the temperature field and die velocity, so as to build a forging digital twin system for metal flow direction control.

[0036] 3. The present invention adopts a forging die structure with local temperature control, and controls the metal flow direction during forging by locally controlling the temperature of the forging, which greatly improves the shape filling rate of the forging and the forming quality of the part. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] Figure 1 It is a schematic diagram of the real-time intelligent regulation principle of the metal flow direction during forging in an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0038] In order to make the objectives, technical solutions and advantages of the present invention clearer and more understandable, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention. In addition, the technical features involved in the various embodiments of the present invention described below can be combined with each other as long as they do not conflict with each other.

[0039] A method for real-time intelligent regulation of metal flow direction during forging provided by an embodiment of the present invention, as Figure 1 shown, during the hot forging production process of a servo press, the collected dynamic signals and static process parameters are input into a digital twin system for online monitoring of the cavity filling situation, and finally the temperature field control strategy in the forging die with local temperature control and the speed control strategy in the servo press are output to control the metal flow direction during hot forging and promote the filling of the die cavity.

[0040] Specifically, it includes the following steps:

[0041] S1. Establishment of the digital twin system

[0042] Forming physical data is obtained through a large number of hot simulation experiments, finite element simulations, and hot forging experiments. Dynamic signals such as ultrasonic images, temperature field distributions, forming forces, displacements, etc., and static process parameters such as equipment models, blank compositions, and forging shapes are collected by an ultrasonic detection system, a temperature measurement system, and a force / displacement sensor of a press in the data acquisition layer and stored in a database; then the multi-source data collected during the hot forging process is subjected to feature extraction and then incorporated into the digital twin system to establish and train a temperature field reconstruction model, a metal plastic flow prediction model, and a forging defect prediction model; further, a forging digital twin system for metal flow direction control is built.

[0043] Preferably, for establishing the temperature field reconstruction model:

[0044] Through finite element simulation of the forging process with different temperature gradients, data extraction is carried out on the simulation results, discrete temperature points corresponding to the temperature measurement positions in the controllable local temperature-controlled forging die are extracted, as well as the overall temperature field distribution. Taking the temperature and coordinates of the discrete points as inputs, and the temperature field distribution as the output and verification, a BP neural network is used to establish a model for reconstructing the temperature field of the forging through local temperature points.

[0045] Preferably, for establishing a metal plastic flow prediction model:

[0046] A series of temperatures and strain rates are selected for hot simulation experiments near reasonable hot forging conditions. Taking 6082 aluminum alloy as an example for illustration: the deformation temperatures (350, 400, 450, 500 °C) and strain rates (0.01, 0.1, 1, 10 s -1 ) are selected, and a series of stress-strain curves are obtained through hot simulation compression experiments on a hot simulation testing machine. According to the results of the hot simulation experiments, a stress-strain constitutive model is established. In this embodiment, the Arrhenius constitutive model of hyperbolic sine is selected:

[0047]

[0048]

[0049] In the formula, σ, R, T, Q are the strain rate, flow stress, gas constant, absolute temperature and activation energy respectively, and A is a structure factor proportional to the density of thermally activated positions in the rate control mechanism at high stress levels.

[0050] According to the established constitutive model, a series of hot forging conditions with different overall strain rates and local temperature distributions are simulated through finite element method. Data extraction is carried out on the finite element simulation. Taking the local temperature distribution and the overall strain rate as inputs, and the metal plastic flow field as the output, a long short-term memory artificial neural network (LSTM) is used to establish a metal plastic flow prediction model based on the local temperature distribution and the overall strain rate.

[0051] Preferably, for establishing a forging defect prediction model:

[0052] A hot processing map is established according to the hot simulation experiment. The power dissipation rate η in the hot processing map can be expressed as:

[0053]

[0054] Among them, J, P, G, m represent the dissipation co-variance, input energy, dissipation amount and strain rate sensitivity factor respectively; the instability criterion in the hot processing map adopts the Prasad instability criterion, and the instability value is:

[0055]

[0056] The instability values are all negative; the power dissipation rate and instability coefficient output by the hot processing map are used to judge whether there are forging defects in the production process.

[0057] S2. Real-time regulation of metal flow direction

[0058] (1) During the hot forging production process of the servo press, dynamic signals such as ultrasonic images, local die temperature, forming force, displacement, etc. collected in real time by the ultrasonic detection system, temperature measurement system, and force / displacement sensors of the press, as well as static process parameters such as equipment model, blank composition, and forging shape, are input into the digital twin system.

[0059] (2) Based on the forming physical data, static parameters, and dynamic signals, the digital twin system reconstructs the temperature field in the forging in real time, and then simulates the plastic flow of the metal in the forging die and the working state of the equipment.

[0060] (3) Predict the possible defects during the forging process, make real-time decisions on temperature field control and slider movement speed control, that is, formulate local heating schemes and servo press movement speeds, so as to control the metal flow direction through force / heat regulation during the forging process.

[0061] (4) Repeat steps (1)-(3), that is, take the data in each forging production as new input data and incorporate it into the digital twin system to update the temperature field control and slider movement speed of the digital twin.

[0062] Furthermore, in step (3), with the optimization objectives of small filling rate deviation of each cavity of the current die, high final overall filling rate, and low forging defects, the genetic algorithm (GA) is used to determine the local temperature field control strategy and slider speed control strategy. Specifically, the metal plastic flow prediction model is used to simulate the metal plastic flow field and the working state of the equipment, and then predict the filling rate of each cavity of the current die and the final overall filling rate; the metal plastic flow field is input into the forging defect prediction model to predict the possible defects during the forging process. In addition, for the three optimization objectives of small filling rate deviation of each cavity of the current die, high final overall filling rate, and low forging defects, they can be weighted according to needs as the optimization objectives in the genetic algorithm.

[0063] Furthermore, the input data needs to be processed. Specifically, a convolutional neural network is used to extract features from the ultrasonic images collected in real time, reconstruct the shape of the current blank, obtain the cavity filling situation during the current forging process by subtracting the shape of the current blank from the die shape, and obtain the strain rate distribution of the blank by taking the time difference of the blank.

[0064] In addition, a heating system composed of a series of electric heating wires and a temperature measuring system composed of a series of thermocouples are built inside the locally temperature-controlled forging die. The electric heating wires and the thermocouples are arranged near the die cavity to achieve the purposes of local temperature control and temperature field monitoring.

[0065] Preferably, nickel-chromium electric heating wires are used in the heating system, and the maximum working temperature is 1100 °C, which can be formed within a relatively wide temperature range.

[0066] Preferably, the ultrasonic detection system adopted is a phased array ultrasonic detection system, which can obtain a two-dimensional image of the material filling condition and realize the accurate prediction of the metal plastic flow field.

[0067] The present invention constructs a digital twin system for regulating the metal flow direction during forging by collecting the dynamic information and static process parameters during the forging process, realizes the real-time monitoring of the cavity filling condition and the prediction of the material flow during the forging process, realizes the intelligent real-time regulation of the metal flow direction during the forging process, achieves the monitoring and management of the entire production cycle, greatly improves the production efficiency and part quality, and can be applied to the forging processes of materials such as aluminum alloy, titanium alloy, magnesium alloy, and steel.

[0068] It is easy for those skilled in the art to understand that the above are only preferred embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent replacements, and improvements made within the spirit and principle of the present invention shall be included within the protection scope of the present invention.

Claims

1. A real-time intelligent control method for metal flow direction during forging, characterized in that, It includes the digital twin system establishment stage and the real-time metal flow regulation stage, where: Digital twin system establishment stage: Obtain the dynamic signals and static process parameters in the hot forging production process, and determine the corresponding forming physical data to form a data set; Based on the data set, establish a digital twin system. Specifically, reconstruct the temperature field in the forging through the digital twin system, and then simulate the plastic flow of the metal in the forging die to predict the defects in the forging process; Real-time metal flow regulation stage: Obtain the dynamic signals and static process parameters in the hot forging production process in real time, and input them into the digital twin system to predict the filling rate of each cavity of the current die, the final overall filling rate, and the forging defects; Taking the small deviation of the filling rate of each cavity of the current die, the high final overall filling rate, and the low forging defects as the optimization objectives, and the local temperature field and slider speed as the optimization variables, use the genetic algorithm to obtain the control strategy, and adjust the local temperature field and slider speed of the hot forging production in real time according to the control strategy to achieve the real-time regulation of the metal flow in the forging production process.

2. The real-time intelligent control method for metal flow direction during forging according to claim 1, characterized in that, Based on the data set, establish a digital twin system, specifically: In the digital twin system, there are a temperature field reconstruction model, a metal plastic flow prediction model, and a forging defect prediction model, which are respectively used to reconstruct the temperature field in the forging, simulate the plastic flow of the metal in the forging die, and predict the defects in the forging process; construct each model through the data set to obtain the digital twin system used in the real-time metal flow regulation stage.

3. The real-time intelligent control method for metal flow direction during forging according to claim 2, characterized in that, Construct the temperature field reconstruction model, specifically: Conduct finite element simulations on the forging processes with different temperature gradients to obtain the discrete point temperatures at the temperature measurement positions in the controllable local temperature-controlled forging die and the overall temperature field distribution; Taking the discrete point temperatures and coordinates as inputs and the temperature field distribution as the output, train the temperature field reconstruction model constructed by the BP neural network to obtain the trained temperature field reconstruction model.

4. The real-time intelligent control method for metal flow direction during forging according to claim 2, characterized in that, Construct the metal plastic flow prediction model, specifically: Conduct hot simulation experiments with different deformation temperatures and strain rates, establish a stress-strain constitutive model based on the experimental results, and then based on the constitutive model, simulate the hot forging conditions with different overall strain rates and local temperature distributions through finite element simulation to obtain the metal plastic flow field; Taking the overall strain rate and local temperature distribution as inputs and the metal plastic flow field as the output, train the metal plastic flow prediction model constructed by the long short-term memory artificial neural network to obtain the trained metal plastic flow prediction model.

5. The real-time intelligent control method for metal flow direction during forging according to claim 4, characterized in that, Construct the forging defect prediction model, specifically: Establish a hot processing diagram through the hot simulation experimental data, and then based on the hot processing diagram, obtain the power dissipation rate and instability coefficient of the hot processing output through the metal plastic flow field and the temperature field, so as to judge whether there are forging defects.

6. The real-time intelligent control method for metal flow direction during forging according to claim 2, characterized in that, In the real-time metal flow regulation stage, predict the filling rate of each cavity of the current die, the final overall filling rate, and the forging defects through the digital twin system, specifically: Input the dynamic signals and static process parameters into the temperature field reconstruction model to obtain the real-time reconstructed temperature field distribution; then input the temperature field distribution into the metal plastic flow prediction model to simulate the metal plastic flow field and the working state of the equipment, and further predict the filling rate of each cavity and the final overall filling rate of the current die; input the metal plastic flow field and the temperature field into the forging defect prediction model to predict the possible defects in the forging process.

7. The real-time intelligent control method for metal flow direction during forging according to claim 1, characterized in that, The dynamic signals include ultrasonic images, local die temperature, forming force, and displacement; the static process parameters include equipment model, blank composition, and forging shape.

8. The real-time intelligent control method for metal flow direction during forging according to any one of claims 1-7, characterized in that, The local temperature field adjustment of hot forging production is realized through a locally temperature-controlled forging die; the locally temperature-controlled forging die includes a heating system with multiple electric heating wires and a temperature measurement system with multiple thermocouples; the electric heating wires and thermocouples are arranged near the die cavity to achieve local temperature control and temperature field monitoring.

9. A real-time intelligent control system for metal flow direction during forging, characterized in that, It includes a processor, and the processor is used to execute the real-time intelligent control method for metal flow direction during forging as described in any one of claims 1-8.

10. A computer-readable storage medium, on which a computer program is stored, characterized in that, When the computer program is executed by the processor, it realizes the real-time intelligent control method for metal flow direction during forging as described in any one of claims 1-8.

Images