CN-121545643-B - TC11 titanium alloy thick-wall pipe diameter forging process parameter optimizing method

CN121545643BCN 121545643 BCN121545643 BCN 121545643BCN-121545643-B

Abstract

The invention discloses a TC11 titanium alloy thick-wall pipe diameter forging process parameter optimizing method, which relates to the technical field of metal plastic forming process optimization, and comprises the steps of definitely defining the problem of tissue defect caused by uneven deformation energy of a specific pipe blank in diameter forging and optimizing targets; the method comprises the steps of constructing a simulation model to calculate the forming size and internal deformation energy distribution under different process parameters, training a neural network proxy model embedded with deformation energy uniformity optimization targets, driving a multi-target optimization algorithm by using the model, automatically searching for the optimal process parameter combination, and selecting and outputting final optimal process parameters from the optimization results. According to the invention, the quantitative index reflecting the uniformity of deformation energy distribution is embedded into the process optimization model, so that the deformation energy distribution in the material can be cooperatively optimized while the macroscopic size is optimized, and the problem that the hidden factor affecting the final organization performance of the component is difficult to effectively evaluate and actively regulate in the process design stage in the conventional method is solved.

Inventors

FENG HAO
GAO YUAN
Weng Qiuwu
CHANG LIJUAN

Assignees

宝鸡宝钛精密锻造有限公司

Dates

Publication Date: 20260505
Application Date: 20260120

Claims (5)

The optimizing method for the thick-wall pipe diameter forging technological parameters of the TC11 titanium alloy is characterized by comprising the following steps of: Determining the specific application scene and the corresponding microstructure defect problem of the TC11 titanium alloy thick-wall pipe diameter forging forming process, and defining the optimization problem comprising technological parameter variables, optimization targets and constraint conditions according to the specific application scene and the corresponding microstructure defect problem; Establishing a high-fidelity multi-physical-field finite element simulation model of the radial forging forming process in the application scene, and collecting simulation result data under different process parameter combinations based on the model, wherein the simulation result data at least comprises final forging outer diameter size and equivalent strain energy density historical data of a plurality of integral points distributed along the radial direction of the cross section of the tube blank at the final forging moment; In the construction process of the neural network proxy model, a deformation energy distribution balance degree calculating layer is embedded, and automatically calculates a balance degree index for quantitatively evaluating radial energy distribution uniformity based on the predicted radial deformation energy density distribution, and the balance degree index is used as one of optimization targets to be integrated into a training loss function of the neural network so as to guide the model to learn the internal rule of the process parameters capable of improving the balance degree index; Based on an improved multi-objective optimization algorithm, iteratively optimizing the technological parameter variables by taking the trained neural network agent model as an fitness evaluation function, wherein the optimization algorithm applies self-adaptive selection pressure to an optimization objective dimension corresponding to the balance index in the evolution process so as to obtain a pareto optimal solution set on two objectives of final forging size precision and deformation energy distribution balance; and selecting a final optimal process parameter set from the pareto optimal solution set and outputting the final optimal process parameter set.
2. The optimizing method of TC11 titanium alloy thick-wall pipe diameter forging technological parameters according to claim 1, wherein the specific application scene is the diameter forging forming of TC11 titanium alloy thick-wall pipes with specific specifications for high-pressure compressor rotors of aeroengines, and the microstructure defect problem is that the deformation energy of a pipe blank core area is excessively concentrated in the diameter forging process, so that abnormal growth of crystal grains occurs in the area during subsequent heat treatment.
3. The method for optimizing parameters of a thick-wall pipe diameter forging process for TC11 titanium alloy according to claim 1, wherein the calculation process of the balance index calculated by the balance calculation layer for deformation energy distribution comprises the steps of calculating an average value of predicted radial deformation energy density distribution, normalizing the distribution based on the average value, and evaluating the dispersion degree of the normalized distribution by adopting high-order statistics, wherein the high-order statistics are used for amplifying the contribution of local sharp peaks in the distribution so as to enhance the sensitivity of the index to deformation energy concentration phenomenon.
4. The method for optimizing parameters of a TC11 titanium alloy thick-wall pipe diameter forging process according to claim 1, wherein the physical reinforced neural network proxy model is a fully-connected neural network with a shared hidden layer and a double-output branch structure, a first output branch is used for predicting final forging outer diameter size, a second output branch is used for predicting radial deformation energy density distribution vector, and the training loss function is composed of three parts, namely a prediction error term of the final outer diameter size, an overall prediction error term of the deformation energy density distribution vector and an optimization term directly based on the balance index.
5. The method for optimizing parameters of a thick-wall pipe diameter forging process of a TC11 titanium alloy according to claim 1, wherein the improved multi-objective optimization algorithm adopts a framework of a non-dominant ranking genetic algorithm with elite strategy, and the improvement is that, when individual crowding distance calculation is carried out, an adaptively-changed weight factor is added to a target dimension component corresponding to the balance index, the weight factor is increased along with the increase of algorithm iteration algebra, so as to strengthen the optimization of the balance of deformation energy distribution in the later period of optimization.

Description

TC11 titanium alloy thick-wall pipe diameter forging process parameter optimizing method Technical Field The invention belongs to the technical field of metal plastic forming process optimization, and particularly relates to a TC11 titanium alloy thick-wall pipe diameter forging process parameter optimizing method. Background The TC11 titanium alloy is a martensite type alpha+beta two-phase titanium alloy with good comprehensive performance, is alloyed by adding elements such as aluminum, tin, zirconium, molybdenum and the like, has higher specific strength and lasting strength at high temperature, has good heat stability and creep resistance, is a key material for manufacturing hot end parts such as aeroengine compressor discs, blades and the like, and has the characteristics of large deformation resistance and relatively narrow plastic forming window in the hot working process. Radial forging, which is fully called radial forging, is a precise forming process for synchronously or alternately forging a rotating or fixed blank by using a high-speed hammer head in multiple directions (usually two or four symmetrical directions), and the process can effectively improve the stress state inside a material and improve the forging compression ratio by partial continuous radial compression deformation, and is particularly suitable for forming thick-wall pipes, stepped shaft parts and difficult-to-deform metal materials, and the characteristics of multidirectional forging are favorable for obtaining more compact and uniform internal tissues. When the diameter forging process is optimized, the prior art method mainly focuses on controlling the macroscopic dimensional accuracy, the surface quality or the overall mechanical property of the forging by adjusting the process parameters such as temperature, deformation and the like. Such optimization usually depends on the combination of finite element simulation and an optimization algorithm, and most of optimization targets are geometric dimension deviation minimization which can be directly measured such as outer diameter, inner diameter and the like, or extreme value control of single physical field quantity such as equivalent stress, strain field and the like after forging. However, for critical components such as aero-engine rotors, which have extremely high requirements on the uniformity of the internal structure of the material, the radial forging forming quality is not only dependent on the macroscopic size, but also limited by the uniformity of the internal microstructure of the material, and the latter is closely related to the distribution characteristic of deformation energy in the material during forming. At present, in the process design stage, a means for effectively evaluating and actively regulating the distribution uniformity of the deformation energy in the material during the forming process is lacking. The deformation energy distribution is used as a process state quantity, is difficult to directly measure on line, and the traditional post-processing analysis based on finite element simulation is also limited to observing a distribution cloud chart or calculating a statistical variance, and cannot be refined into a quantization index which can be parallel to a macroscopic size target and can be embedded into an automatic optimization flow. Therefore, when the problem that the deformation energy of the core part is concentrated due to the metal flow characteristic of the thick-wall tube blank with specific specification is faced, the prior optimization method is difficult to prevent and inhibit the hidden defects in advance in the parameter optimizing process. This limitation results in process development that still relies to some extent on empirical trial and error. Although the existing optimization method can effectively converge to a parameter solution meeting the macroscopic size requirement, the solution may not be optimal in terms of deformation energy distribution uniformity, and may even be in a parameter interval that is prone to induce structural anomalies (such as coarse local grains) in subsequent heat treatment. Accordingly, the following scheme is proposed for the above problems. Disclosure of Invention The invention aims to provide a TC11 titanium alloy thick-wall pipe diameter forging process parameter optimizing method, which can be used for optimizing macroscopic dimensions and simultaneously optimizing the internal deformation energy distribution of a material in a cooperative manner by embedding a quantization index reflecting deformation energy distribution uniformity into a process optimizing model, and solves the problem that the conventional method is difficult to effectively evaluate and actively regulate and control the hidden factors influencing the final structural performance of a component in a process design stage. In order to solve the technical problems, the invention is realized by the following technical scheme: th