Solder Joint Reliability Risk Estimation by AI-Assisted Simulation Framework with Genetic Algorithm to Optimize the Initial Parameters for AI Models

Solder joint fatigue is one of the critical failure modes in ball-grid array packaging. Because the reliability test is time-consuming and geometrical/material nonlinearities are required for the physics-driven model, the AI-assisted simulation framework is developed to establish the risk estimation capability against the design and process parameters. Due to the time-dependent and nonlinear characteristics of the solder joint fatigue failure, this research follows the AI-assisted simulation framework and builds the non-sequential artificial neural network (ANN) and sequential recurrent neural network (RNN) architectures. Both are investigated to understand their capability of abstracting the time-dependent solder joint fatigue knowledge from the dataset. Moreover, this research applies the genetic algorithm (GA) optimization to decrease the influence of the initial guessings, including the weightings and bias of the neural network architectures. In this research, two GA optimizers are developed, including the “back-to-original” and “progressing” ones. Moreover, we apply the principal component analysis (PCA) to the GA optimization results to obtain the PCA gene. The prediction error of all neural network models is within 0.15% under GA optimized PCA gene. There is no clear statistical evidence that RNN is better than ANN in the wafer level chip-scaled packaging (WLCSP) solder joint reliability risk estimation when the GA optimizer is applied to minimize the impact of the initial AI model. Hence, a stable optimization with a broad design domain can be realized by an ANN model with a faster training speed than RNN, even though solder fatigue is a time-dependent mechanical behavior.     

Alloy Partitioning Effect on Strength and Toughness of κ-Carbide Strengthened Steels

Alloy partitioning during heat treatment in a lightweight precipitation hardened steel was investigated using transmission electron microscopy and atom probe tomography. The mechanical properties are discussed as a function of the effect of solution treatment temperature and aging time, giving rise to variations in chemical modulation. A wrought lightweight steel alloy with a nominal composition of Fe-30Mn-9Al-1Si-1C-0.5Mo (wt. %) was solution-treated between 1173–1273 K and aged at 773 K. Lower solution treatment temperatures retained a finer grain size and accelerated age hardening response that also produced an improved work hardening behavior with a tensile strength of −1460 MPa at 0.4 true strain. Atom probe tomography indicated these conditions also had reduced modulation in the Si and Al content due to the reduced aging time preventing silicon from diffusing out of the κ-carbide into the austenite. This work provides the framework for heat-treating lightweight, age hardenable steels with high strength and improved energy absorption.     

Accurate Estimation of Yield Strength and Ultimate Tensile Strength through Instrumented Indentation Testing and Chemical Composition Testing

Federal rule changes governing natural gas pipelines have made non-destructive techniques, such as instrumented indentation testing (IIT), an attractive alternative to destructive tests for verifying properties of steel pipeline segments that lack traceable records. Ongoing work from Pacific Gas and Electric Company’s (PG&E) materials verification program indicates that IIT measurements may be enhanced by incorporating chemical composition data. This paper presents data from PG&E’s large-scale IIT program that demonstrates the predictive capabilities of IIT and chemical composition data, with particular emphasis given to differences between ultimate tensile strength (UTS) and yield strength (YS). For this study, over 80 segments of line pipe were evaluated through tensile testing, IIT, and compositional testing by optical emission spectroscopy (OES) and laboratory combustion. IIT measurements of UTS were, generally, in better agreement with destructive tensile data than YS and exhibited about half as much variability as YS measurements on the same sample. The root-mean squared error for IIT measurements of UTS and YS, respectively, were 27 MPa (3.9 ksi) and 43 MPa (6.2 ksi). Next, a machine learning model was trained to estimate YS and UTS by combining IIT with chemical composition data. The agreement between the model’s estimated UTS and tensile UTS values was only slightly better than the IIT-only measurements, with an RMSE of 21 MPa (3.1 ksi). However, the YS estimates showed much greater improvement with an improved RMSE of 27 MPa (3.9 ksi). The experimental, mechanical, and metallurgical factors that contributed to IIT’s ability to consistently determine destructive UTS, and the differences in its interaction with composition as compared to YS, are discussed herein.     

Back-Propagation Neural Network Optimized by K-Fold Cross-Validation for Prediction of Torsional Strength of Reinforced Concrete Beam

Due to the limitation of sample size in predicting the torsional strength of Reinforced Concrete (RC) beams, this paper aims to discuss the feasibility of employing a novel machine learning approach with K-fold cross-validation in a small sample range, which combines the advantages of a Genetic Algorithm (GA) and a Neural Network (NN) to predict the torsional strength of RC beams. This research study not only utilizes the application of a Back Propagation (BP) neural network and the Gene Algorithm-Back Propagation (GA-BP) neural network in the prediction of the torsional strength of the RC beam, but it also investigates neural network parameter optimization, including connection weights and thresholds, using K-fold cross-validation. The root mean square error (RMSE), mean absolute error (MAE), mean square error (MSE), mean absolute percentage error (MAPE) and correlation coefficient (R²) are among the evaluation metrics used to assess the performance of the trained model. To elaborate on the superiority of the proposed network models in predicting the torsional strength of RC beams, a parametric study is conducted by comparing the proposed model to three commonly used empirical formulae from existing design codes. The comparative findings of this research study demonstrate that the performance of the BP neural network is highly similar to that of design codes; however, its accuracy is inadequate. After improving the weights and thresholds by k-fold cross-validation and GA, the prediction of the BP neural network shows higher consistency with the actual measured values. The outcome of this study can be used as a theoretical reference for the optimal design of RC beams in practical applications.     

The Effect of Quenching and Partitioning (Q&P) Heat Treatment on the Microstructure and Mechanical Properties of High Boron Steel

High boron steel is prone to brittle failure due to the boride distributed in it with net-like or fishbone morphology, which limit its applications. The Quenching and Partitioning (Q&P) heat treatment is a promising process to produce martensitic steel with excellent mechanical properties, especially high toughness by increasing the volume fraction of retained austensite (RA) in the martensitic matrix. In this work, the Q&P heat treatment is used to improve the inherent defect of insufficient toughness of high boron steel, and the effect mechanism of this process on microstructure transformation and the change of mechanical properties of the steel has also been investigated. The high boron steel as-casted is composed of martensite, retained austensite (RA) and eutectic borides. A proper quenching and partitioning heat treatment leads to a significant change of the microstructure and mechanical properties of the steel. The net-like and fishbone-like boride is partially broken and spheroidized. The volume fraction of RA increases from 10% in the as-cast condition to 19%, and its morphology also changes from blocky to film-like. Although the macro-hardness has slightly reduced, the toughness is significantly increased up to 7.5 J·cm⁻², and the wear resistance is also improved.     

Laser Powder Bed Fusion of Ti-6Al-2Sn-4Zr-6Mo Alloy and Properties Prediction Using Deep Learning Approaches

Ti-6Al-2Sn-4Zr-6Mo is one of the most important titanium alloys characterised by its high strength, fatigue, and toughness properties, making it a popular material for aerospace and biomedical applications. However, no studies have been reported on processing this alloy using laser powder bed fusion. In this paper, a deep learning neural network (DLNN) was introduced to rationalise and predict the densification and hardness due to Laser Powder Bed Fusion of Ti-6Al-2Sn-4Zr-6Mo alloy. The process optimisation results showed that near-full densification is achieved in Ti-6Al-2Sn-4Zr-6Mo alloy samples fabricated using an energy density of 77–113 J/mm³. Furthermore, the hardness of the builds was found to increase with increasing the laser energy density. Porosity and the hardness measurements were found to be sensitive to the island size, especially at high energy density. Hot isostatic pressing (HIP) was able to eliminate the porosity, increase the hardness, and achieve the desirable α and β phases. The developed model was validated and used to produce process maps. The trained deep learning neural network model showed the highest accuracy with a mean percentage error of 3% and 0.2% for the porosity and hardness. The results showed that deep learning neural networks could be an efficient tool for predicting materials properties using small data.     

Tensile Properties of 21-6-9 Austenitic Stainless Steel Built Using Laser Powder-Bed Fusion

Alloy 21-6-9 is an austenitic stainless steel with high strength, thermal stability at high temperatures, and retained toughness at cryogenic temperatures. This type of steel has been used for aerospace applications for decades, using traditional manufacturing processes. However, limited research has been conducted on this alloy manufactured using laser powder-bed fusion (LPBF). Therefore, in this work, a design of experiment (DOE) was performed to obtain optimized process parameters with regard to low porosity. Once the optimized parameters were established, horizontal and vertical blanks were built to investigate the mechanical properties and potential anisotropic behavior. As this alloy is exposed to elevated temperatures in industrial applications, the effect of elevated temperatures (room temperature and 750 °C) on the tensile properties was investigated. In this work, it was shown that alloy 21-6-9 could be built successfully using LPBF, with good properties and a density of 99.7%, having an ultimate tensile strength of 825 MPa, with an elongation of 41%, and without any significant anisotropic behavior.     

Swin–UNet++: A Nested Swin Transformer Architecture for Location Identification and Morphology Segmentation of Dimples on 2.25Cr1Mo0.25V Fractured Surface

The precise identification of micro-features on 2.25Cr1Mo0.25V steel is of great significance for understanding the mechanism of hydrogen embrittlement (HE) and evaluating the alloy’s properties of HE resistance. Presently, the convolution neural network (CNN) of deep learning is widely applied in the micro-features identification of alloy. However, with the development of the transformer in image recognition, the transformer-based neural network performs better on the learning of global and long-range semantic information than CNN and achieves higher prediction accuracy. In this work, a new transformer-based neural network model Swin–UNet++ was proposed. Specifically, the architecture of the decoder was redesigned to more precisely detect and identify the micro-feature with complex morphology (i.e., dimples) of 2.25Cr1Mo0.25V steel fracture surface. Swin–UNet++ and other segmentation models performed state-of-the-art (SOTA) were compared on the dimple dataset constructed in this work, which consists of 830 dimple scanning electron microscopy (SEM) images on 2.25Cr1Mo0.25V steel fracture surface. The segmentation results show Swin–UNet++ not only realizes the accurate identification of dimples but displays a much higher prediction accuracy and stronger robustness than Swin–Unet and UNet. Moreover, efforts from this work will also provide an important reference value to the identification of other micro-features with complex morphologies.     

Application of BP Artificial Neural Network in Preparation of Ni–W Graded Coatings

The internal stress difference between soft-ductile aluminum alloy substrate and hard-brittle Ni–W alloy coating will cause stress concentration, thus leading to the problem of poor bonding force. Herein, this work prepared the Ni–W graded coating on aluminum alloy matrix by the pulse electrodeposition method in order to solve the mechanical mismatch problem between substrate and coatings. More importantly, a backward propagation (BP) neural network was applied to efficiently optimize the pulse electrodeposition process of Ni–W graded coating. The SEM, EDS, XRD, Vickers hardness tester and Weighing scales are used to analyze the micromorphology, chemical element, phase composition, and micro hardness as well as oxidation weight increase, respectively. The results show that the optimal process conditions with BP neural network are as follows: the bath temperature is 30 °C, current density is 15 mA/cm² and duty cycle is 0.3. The predicted value of the model agrees well with the experimental value curve, the relative error is minor. The maximum error is less than 3%, and the correlation coefficient is 0.9996. The Ni–W graded coating prepared by BP neural network shows good bonding with the substrate which has flat and smooth interface. The thickness of the coating is about 136 μm, which slows down the oxidation of the substrate and plays an effective role in protecting the substrate.     

Structure and Mechanical Properties of Shipbuilding Steel Obtained by Direct Laser Deposition and Cold Rolling

The study of the formation of microstructural features of low-alloy bainite-martensitic steel 09CrNi2MoCu are of particular interest in additive technologies. In this paper, we present a study of cold-rolled samples after direct laser deposition (DLD). We investigated deposited samples after cold plastic deformation with different degrees of deformation compression (50, 60 and 70%) of samples from steel 09CrNi2MoCu. The microstructure and mechanical properties of samples in the initial state and after heat treatment (HT) were analyzed and compared with the samples obtained after cold rolling. The effect on static tensile strength and impact toughness at −40 °C in the initial state and after cold rolling was investigated. The mechanical properties and characteristics of fracture in different directions were determined. Optimal modes and the degree of cold rolling deformation compression required to obtain balanced mechanical properties of samples obtained by additive method were determined. The influence of structural components and martensitic-austenitic phase on the microhardness and mechanical properties of the obtained samples was determined.     

Overview of HSS Steel Grades Development and Study of Reheating Condition Effects on Austenite Grain Size Changes

This review paper concerns the development of the chemical compositions and controlled processes of rolling and cooling steels to increase their mechanical properties and reduce weight and production costs. The paper analyzes the basic differences among high-strength steel (HSS), advanced high-strength steel (AHSS) and ultra-high-strength steel (UHSS) depending on differences in their final microstructural components, chemical composition, alloying elements and strengthening contributions to determine strength and mechanical properties. HSS is characterized by a final single-phase structure with reduced perlite content, while AHSS has a final structure of two-phase to multiphase. UHSS is characterized by a single-phase or multiphase structure. The yield strength of the steels have the following value intervals: HSS, 180–550 MPa; AHSS, 260–900 MPa; UHSS, 600–960 MPa. In addition to strength properties, the ductility of these steel grades is also an important parameter. AHSS steel has the best ductility, followed by HSS and UHSS. Within the HSS steel group, high-strength low-alloy (HSLA) steel represents a special subgroup characterized by the use of microalloying elements for special strength and plastic properties. An important parameter determining the strength properties of these steels is the grain-size diameter of the final structure, which depends on the processing conditions of the previous austenitic structure. The influence of reheating temperatures (T_Reh) and the holding time at the reheating temperature (t_Reh) of C–Mn–Nb–V HSLA steel was investigated in detail. Mathematical equations describing changes in the diameter of austenite grain size (d_γ), depending on reheating temperature and holding time, were derived by the authors. The coordinates of the point where normal grain growth turned abnormal was determined. These coordinates for testing steel are the reheating conditions T_Reh = 1060 °C, t_Reh = 1800 s at the diameter of austenite grain size d_γ = 100 μm.     

The Effect of Zr Addition on Melting Temperature,Microstructure, Recrystallization and Mechanical Properties of a Cantor High Entropy Alloy

The effect of Zr addition on the melting temperature of the CoCrFeMnNi High Entropy Alloy (HEA), known as the “Cantor’s Alloy”, is investigated, together with its micro-structure, mechanical properties and thermomechanical recrystallization process. The base and Zr-modified alloys are obtained by vacuum induction melting of mechanically pre-alloyed powders. Raw materials are then cold rolled and annealed. recrystallization occurred during the heat treatment of the cold-rolled HEA. The alloys are characterized by X-ray diffraction, electron microscopy, thermal analyses, mechanical spectroscopy and indentation measures. The main advantages of Zr addition are: (1) a fast vacuum induction melting process; (2) the lower melting temperature, due to Zr eutectics formation with all the Cantor’s alloy elements; (3) the good chemical alloy homogeneity; and (4) the mechanical properties improvement of re-crystallized grains with a coherent structure. The crystallographic lattice of both alloys results in FCC. The Zr-modified HEA presents a higher recrystallization temperature and smaller grain size after recrystallization with respect to the Cantor’s alloy, with precipitation of a coherent second phase, which enhances the alloy hardness and strength.     

Wear-Resistance Improvement of 65Mn Low-Alloy Steel through Adjusting Grain Refinement by Cyclic Heat Treatment

Refined microstructures achieved by cyclic heat treatment significantly contribute to improving the wear resistance of steels. To acquire the refined microstructures of 65Mn low-alloy steel, first, the specimens were solid solution-treated; then, they were subjected to cyclic heat treatment at cyclic quenching temperatures of 790–870 °C and quenching times of 1–4 with a fixed holding time of 5 min. The mechanical properties of 65Mn low-alloy steel in terms of hardness, tensile strength, elongation and wear resistance were characterized. Afterwards, the effect of cyclic heat treatment on microstructure evolution and the relationships between grain refinement and mechanical properties’ improvement were discussed. The results show that the average grain size firstly decreased and then increased with the increase in the quenching temperature. Hardness increased with grain refinement when the temperature was lower than 830 °C. Once the temperature exceeded 830 °C, hardness increased with the temperature increase owing to the enrichment of carbon content in the martensite. With the increase in cyclic quenching times, hardness continuously increased with grain refinement strengthening. In addition, both tensile strength and elongation could be significantly improved through grain refinement. The relationships among wear loss, hardness and average grain size showed that wear resistance was affected by the synthesis reaction of grain refinement and hardness. Higher hardness and refined grain size contributed to improving the wear resistance of 65Mn low-alloy steel.     

The Influence of Annealing Temperature on the Morphology of Structures and the Mechanical Properties of Prequenching—Quenching and Partitioning Steel

In this study, we comparatively study the microstructures and mechanical properties of prequenching—quenching and partitioning (QQ&P) and traditional Q&P samples at different annealing temperatures (intercritical annealing temperatures). When the annealing temperature is 780 °C, the ferrite and retained austenite in QQ&P samples with lath and blocky morphologies. The lath retained austenite is mainly distributed along the lath ferrite. As the annealing temperature increases, the lath ferrite recrystallizes and gradually grows into the blocky (equiaxed) shape, leading to a decrease in the lath retained austenite content. When the annealing temperature increases to 870 °C, the ferrite content decreases significantly, and the retained austenite is mainly blocky and thin film, distributed at the boundaries of prior austenite grains and between martensite laths, respectively. Different from QQ&P samples, the ferrite and retained austenite in Q&P samples are mainly blocky when the annealing temperature is 780 °C or 810 °C. When the annealing temperature is increased to 870 °C, the microstructures of the Q&P sample are basically the same as that of the QQ&P sample. The 780 °C-QQ&P sample and the 810 °C-QQ&P sample have higher total elongation and product of strength and elongations (PSEs) than their counterpart Q&P samples due to the fact that lath ferrite and retained austenite are conducive to carbon diffusion and carbon homogenization in austenite grains, thereby improving the thermal stability and volume fraction of the retained austenite. In addition, the lath structures can release local stress concentration and delay the formation of voids and microcracks. The difference of mechanical properties between QQ&P samples and Q&P samples decreases with the increase in the annealing temperature. The results show that the low annealing temperature combined with prequenching—Q&P heat treatments can significantly improve the elongation and PSE of Q&P steel.     

Evaluation of Steels Susceptibility to Hydrogen Embrittlement: A Thermal Desorption Spectroscopy-Based Approach Coupled with Artificial Neural Network

A novel approach has been developed for quantitative evaluation of the susceptibility of steels and alloys to hydrogen embrittlement. The approach uses a combination of hydrogen thermal desorption spectroscopy (TDS) analysis with recent advances in machine learning technology to develop a regression artificial neural network (ANN) model predicting hydrogen-induced degradation of mechanical properties of steels. We describe the thermal desorption data processing, artificial neural network architecture development, and the learning process beneficial for the accuracy of the developed artificial neural network model. A data augmentation procedure was proposed to increase the diversity of the input data and improve the generalization of the model. The study of the relationship between thermal desorption spectroscopy data and the mechanical properties of steel evidences a strong correlation of their corresponding parameters. A prototype software application based on the developed model is introduced and is openly available. The developed prototype based on TDS analysis coupled with ANN is shown to be a valuable engineering tool for steel characterization and quantitative prediction of the degradation of steel properties caused by hydrogen.     

Effect of Material Heterogeneity on Environmentally Assisted Cracking Growth Rate of Alloy 600 for Safe-End Welded Joints

Environmentally assisted cracking (EAC) is essential in predicting light water reactors’ structural integrity and service life. Alloy 600 (equivalent to Inconel 600) has excellent corrosion resistance and is often used as a welding material in welded joints, but material properties of the alloy are heterogeneous in the welded zone due to the complex welding process. To investigate the EAC crack growth behavior of Alloy 600 for safe-end welded joints, the method taken in this paper concerns the probability prediction of the EAC crack growth rate. It considers the material heterogeneity, combining the film slip-dissolution/oxidation model, and the elastic-plastic finite element method. The strain rate at the crack tip is a unique factor to describe the mechanical state. Still, it is challenging to accurately predict it because of the complicated and heterogeneous material microstructure. In this study, the effects of material heterogeneity on the EAC crack growth behavior are statistically analyzed. The results show that the material heterogeneity of Alloy 600 can not be ignored because it affects the prediction accuracy of the crack growth rate. The randomness of yield strength has the most influence on the EAC growth rate, while Poisson’s ratio has the smallest.     

A Hybrid Approach of ANFIS—Artificial Bee Colony Algorithm for Intelligent Modeling and Optimization of Plasma Arc Cutting on Monel™ 400 Alloy

This paper focusses on a hybrid approach based on genetic algorithm (GA) and an adaptive neuro fuzzy inference system (ANFIS) for modeling the correlation between plasma arc cutting (PAC) parameters and the response characteristics of machined Monel 400 alloy sheets. PAC experiments are performed based on box-behnken design methodology by considering cutting speed, gas pressure, arc current, and stand-off distance as input parameters, and surface roughness (Ra), kerf width (kw), and micro hardness (mh) as response characteristics. GA is efficaciously utilized as the training algorithm to optimize the ANFIS parameters. The training, testing errors, and statistical validation parameter results indicated that the ANFIS learned by GA outperforms in the forecasting of PAC responses compared with the results of multiple linear regression models. Besides that, to obtain the optimal combination PAC parameters, multi-response optimization was performed using a trained ANFIS network coupled with an artificial bee colony algorithm (ABC). The superlative responses, such as R_a of 1.5387 µm, kw of 1.2034 mm, and mh of 176.08, are used to forecast the optimum cutting conditions, such as a cutting speed of 2330.39 mm/min, gas pressure of 3.84 bar, arc current of 45 A, and stand-off distance of 2.01 mm, respectively. Furthermore, the ABC predicted results are validated by conducting confirmatory experiments, and it was found that the error between the predicted and the actual results are lower than 6.38%, indicating the adoptability of the proposed ABC in optimizing real-world complex machining processes.     

Compound Structure–Composition Control on the Mechanical Properties of Selective Laser-Melted Titanium Alloys

In the performance optimization of the additive manufacturing of Ti6Al4V components, conventional control methods have difficulty taking into account the requirements of quality and mechanical properties of components, resulting in insufficient mechanical properties and a small control range. Therefore, combining the advantages of porous structure and alloy composition control, this paper proposed a structure–composition composite control method for selective laser-fused titanium alloy components by coupling the effects of porous structure parameters and boron content on the properties of Ti6Al4V components. Based on the Gibson–Ashby formula, the compression test of porous Ti6Al4V alloy and the tensile test of boron-containing Ti6Al4V alloy were carried out by SLM forming technology. The parameters C and n related to the pore parameters of porous structure were solved by the experimental data, and the analytical relationship between the pore parameters and the mechanical properties of Ti6Al4V alloy was established. The analytical relationship between boron content (t wt%) and mechanical properties of the alloy was established by tensile test. Finally, the Gibson–Ashby formula was used to combine the above analytical relationship, and a composite regulation model of compressive strength was obtained. The results show that the control range of the composite model ranges from 19.46–416.47 MPa, which was 45.53% higher than that obtained by controlling only pore parameters, and performance improved by 42.49%. The mechanical properties of the model are verified and the deviation between calculated values and experimental values was less than 1.3%. Taking aviation rocker arm as an example, the optimized design can improve the strength and reduce the mass of rocker arm by 51.94%. This method provides a theoretical basis for expanding the application of Ti6Al4V additive manufacturing components in aerospace and other fields.     

Effect of Carbon Partition and Precipitation on Wear Resistance of Carburized Layer in Heavy-Duty Gear

The carburizing–quenching–tempering process is generally conducted on heavy-duty gear in order to obtain favorable comprehensive mechanical performance. Different mechanical properties could be produced by carbon partition and precipitation. In this study, the carburizing–quenching–tempering process was carried out on low-carbon alloy steel in order to investigate the influence of microstructure evolution and precipitate transition on mechanical behavior and wear resistance under different carburizing/tempering durations. Favorable comprehensive mechanical property and wear resistance could be obtained in favor of long durations of carburizing/tempering. A fatigue-wear model was proposed to describe fatigue crack evolution and damage mechanism on the basis of wear features.     

Constitutive Model Parameter Identification Based on Optimization Method and Formability Analysis for Ti6Al4V Alloy

Titanium alloy is widely applied in aerospace, medical, shipping and other fields due to its high specific strength and low density. The purpose of this study was to analyze the formability of Ti6Al4V alloys at elevated temperatures. An accurate constitutive model is the basic condition for accurately simulating the plastic forming of materials, and it is an important basis for optimizing the parameters of the hot forging forming process. In this study, the optimization algorithm was used to accurately identify the high-temperature constitutive model parameters of Ti6Al4V titanium alloy, and the hot working diagram was established to optimize the hot forming process parameters. The optimal forming conditions of Ti6Al4V titanium alloy are given. Ti6Al4V alloy was subjected to high-temperature compression tests at 800–1000 °C and at strain rates of 0.01–5 s⁻¹ on a Gleeble-1500D thermal/mechanical simulation machine. Each parameter of the Hansel–Spittel constitutive model was taken as an independent variable, and the accumulated error between the stress calculated by the constitutive model and the stress obtained by experimentation was used as an objective function. Based on response surface methodology, an inverse optimization method for identifying the parameters of the high-temperature constitutive model of Ti6Al4V alloy is proposed in this paper. An orthogonal test design was adopted to obtain sample point data, and a third-order response surface approximate model was established. The genetic algorithm (GA) was applied to reversely optimize the parameters of the constitutive model. To verify the accuracy of the optimized constitutive model, the average absolute relative error (AARE) and correlation coefficient (R) were used to evaluate the reliability of optimized constitutive model. The R value of the model was 0.999, and the AARE value was 0.048, respectively, indicating that the established high-temperature constitutive model for Ti6Al4V alloy has good calculation accuracy. The flow stress behavior of the material could be accurately delineated. Meanwhile, in order to study the formability of Ti6Al4V alloy, the hot processing map of the alloy, based on a dynamic material model, was established in this paper. The optimum hot working domains of the Ti6Al4V alloy were determined within 840–920 °C/0.01–0.049 s⁻¹ and 940–980 °C/0.11–1.65 s⁻¹; the hot processing map was verified in combination with the microstructure, and the fine and equiaxed grains and a large amount of β phase could be found at 850 °C/0.01 s⁻¹.