研究目的
To explore predictive models for residual stresses for single bead tracks over a wide range of process settings for a coaxial powder injection laser cladding process depositing P420 stainless steel powder on low carbon steel plates.
研究成果
The ANFIS model is better than the ANN model for this single bead scenario. Multiple beads will be investigated in ongoing research.
研究不足
The residual stress is dependent on the process settings, and the location within the bead and substrate; hence, a discrete solution could not be readily determined using an ANOVA approach.
1：Experimental Design and Method Selection:
A three-phase approach was taken: (i) experiments to collect geometry and hardness data, (ii) development of finite element simulation models to collect residual stress data, and (iii) generation of predictive models for the induced residual stresses.
2：Sample Selection and Data Sources:
Single pass bead on plate laser cladding tracks of P420 were deposited onto AISI
3：List of Experimental Equipment and Materials:
10 A coaxial deposition head mounted onto a 6 serial axis robot employing a 4kW diode laser was used. Argon gas was employed to protect the melt pool.
4：Experimental Procedures and Operational Workflow:
A comprehensive design of experiments approach was taken to explore five process parameters at five different levels.
5：Data Analysis Methods:
Artificial neural network model and an adaptive neuro fuzzy interference system (ANFIS) model were used for data analysis.
独家科研数据包，助您复现前沿成果，加速创新突破
获取完整内容
