Numerical Simulation of Submerged Arc Weld Overlay Considering Phase Transformation Induced Plasticity

Literature Overview

This paper by Huang Qingchun, Li Chang, Zhang Dacheng, Gao Hexin, Han Xing, and Li Yunfei from Liaoning University of Science and Technology and China Energy Engineering Dongfang Electric Corporation Boiler Project Department, published in the journal Surface Technology in 2021, presents an advanced numerical simulation method for submerged arc weld overlay processes that incorporates the effects of phase transformation induced plasticity (TRIP). The work was supported by multiple funding sources including the National Natural Science Foundation of China (E050402/51105187), the Public Security Bureau Fire Key Laboratory Open Project (KF201704), the Liaoning Provincial Natural Science Foundation (2019ZD0277), and the Liaoning University of Science and Technology Innovation Team Construction Project (601009830-02), reflecting its significance in advancing the computational modeling of weld overlay processes.

Core Technical Content

The study addresses a critical limitation in traditional finite element simulations of weld overlay processes: the neglect of phase transformation effects on the mechanical behavior of the material. During weld overlay, the heat-affected zone and the weld metal undergo phase transformations (austenite to martensite, bainite, or ferrite) that are accompanied by volume changes and induced plastic strains. These transformation-induced plasticity effects can significantly influence the residual stress distribution and the final mechanical properties of the overlay layer.

Phase Transformation Induced Plasticity Model

The authors developed a constitutive model that incorporates the TRIP effect into the finite element analysis. The model is based on the following principles:

Phase transformation kinetics: The transformation from austenite to martensite is modeled using the Koistinen-Marburger equation, which relates the volume fraction of martensite to the temperature difference between the current temperature and the martensite start temperature (Ms).
Transformation-induced strain: The volume expansion associated with the austenite-to-martensite transformation (approximately 1–2%) generates additional plastic strains that are superimposed on the thermal and mechanical strains.
Coupled thermal-mechanical analysis: The model couples the thermal analysis with the mechanical analysis, allowing the phase transformation to influence the stress-strain state and vice versa.

Model Component	Equation/Approach	Key Parameters
Phase transformation kinetics	Koistinen-Marburger equation	Ms temperature, transformation rate
Transformation-induced strain	ε_tr = ε_0 × f_martensite	ε_0 ≈ 0.01–0.02 (volume strain)
Thermal stress	σ_thermal = E × α × ΔT	E (Young's modulus), α (thermal expansion coefficient)
Mechanical stress	σ_mechanical = E × ε_mechanical	Depends on boundary conditions and constraints
Total stress	σ_total = σ_thermal + σ_mechanical + σ_transformation	Superposition of all stress components

Simulation Results and Validation

The simulation results were validated against experimental measurements of residual stresses and microstructures. The key findings include:

Residual stress distribution: The inclusion of TRIP effects increased the predicted residual stresses by 10–25% compared to simulations that neglected phase transformation. The maximum longitudinal residual stress reached 300–400 MPa, which is close to the yield strength of the material.
Microstructure prediction: The model predicted the formation of martensite in the HAZ and the weld metal, with the martensite fraction increasing as the cooling rate increased. The predicted martensite fraction was in good agreement with experimental measurements (within ±10%).
Hardness prediction: The predicted hardness distribution, based on the predicted phase fractions and the rule of mixtures, agreed well with experimental measurements. The surface hardness of the overlay layer was predicted to be 400–600 HV, consistent with experimental values.

Comparison with Traditional Simulation Approaches

Aspect	Traditional Simulation	TRIP-Incorporated Simulation	Improvement
Residual stress prediction	Underestimates by 10–25%	More accurate, within 10% of experiment	Significant improvement in stress prediction
Microstructure prediction	Not included or simplified	Detailed phase transformation modeling	Enables microstructure-based property prediction
Computational cost	Lower	Higher (50–100% increase)	Acceptable for detailed analysis
Applicability	General	Specific to phase-transforming materials	More accurate for steels with phase transformations

Engineering Application and Process Optimization

The advanced simulation method presented in this paper has several important applications in the design and optimization of weld overlay processes:

Process parameter optimization: The model can be used to predict the effects of different welding parameters (current, voltage, travel speed) on the residual stress distribution and the microstructure of the overlay layer. This enables the selection of parameters that minimize residual stresses and optimize the mechanical properties.
Procedure qualification: The model can assist in the qualification of weld overlay procedures by predicting the expected residual stresses and microstructures, reducing the need for extensive experimental testing.
Defect prediction: The model can predict the likelihood of cracking based on the predicted residual stresses and the material's crack susceptibility. This enables the identification of critical process parameters that must be controlled to prevent cracking.
Post-weld treatment optimization: The model can predict the effects of post-weld heat treatment on the residual stress distribution and the microstructure, enabling the optimization of heat treatment parameters.

In practical applications, the TRIP-incorporated simulation method is particularly valuable for the design of overlay layers on high-strength steels and low-alloy steels, where phase transformations are significant and the residual stress state is critical for fatigue and fracture performance.

Study Insights and Practical Value

The incorporation of phase transformation induced plasticity into the numerical simulation of weld overlay processes represents a significant advancement in computational welding mechanics. The model provides a more accurate prediction of the residual stress distribution and the microstructure evolution, which are critical for the design and optimization of weld overlay procedures. The validation of the model against experimental data demonstrates its reliability and applicability to real-world problems. For engineers involved in the design and qualification of weld overlay processes, this study provides a powerful tool for process optimization and defect prediction. The method is particularly valuable for applications where the residual stress state is critical, such as fatigue-critical components and components subjected to high-temperature service.

Conclusion

These five studies collectively represent important contributions to the understanding and optimization of weld overlay processes for dissimilar material systems, wear-resistant steels, and critical engineering components. The common thread across all studies is the recognition that the microstructure, residual stress state, and mechanical properties of the overlay layer are intimately linked to the process parameters and the material system. The numerical simulation approaches presented in Topics 3 and 5 provide powerful tools for process optimization and defect prediction, while the experimental studies in Topics 1, 2, and 4 provide valuable insights into the microstructure-property relationships and the crack formation mechanisms. Together, these studies form a comprehensive foundation for the design, qualification, and application of weld overlay processes in modern engineering practice. The integration of experimental characterization with computational modeling represents the future direction of weld overlay technology, enabling more accurate predictions and more reliable process development.