Ming-Song Chen

Effects of initial δ phase (Ni3Nb) on hot tensile deformation behaviors and material constants of Ni-based superalloy

Abstract Effects of initial δ phase (Ni3Nb) on the hot tensile deformation behaviors and material constants of a Ni-based superalloy were investigated over wide ranges of strain rate and deformation temperature. It is found that the true stress–true strain curves exhibit peak stress at a small strain, and the peak stress increases with the increase of initial δ phase. After the peak stress, initial δ phase promotes the dynamic softening behaviors, resulting in the decreased flow stress. An improved Arrhenius constitutive model is proposed to consider the synthetical effects of initial δ phase, deformation temperature, strain rate, and strain on hot deformation behaviors. In the improved model, material constants are expressed as the functions of the content of initial δ phase and strain. A good agreement between the predicted and measured results indicates that the improved Arrhenius constitutive model can well describe hot deformation behaviors of the studied Ni-based superalloy.

A precise BP neural network-based online model predictive control strategy for die forging hydraulic press machine

The time variance and nonlinearity of forging processes pose great challenges to high-quality production. In this study, a one-step-ahead model predictive control (MPC) strategy based on backpropagation (BP) neural network is proposed for the precise forging processes. Two online updated BP neural networks, predictive neural network (PNN) and control neural network (CNN), are developed to accurately control the die forging hydraulic press machine. The PNN and CNN are utilized to predict the output (the velocity of upper die) and determine the input (the oil pressure of driven cylinders), respectively. The weights of neural networks are initially trained offline and then updated online according to an error backpropagation algorithm. In the proposed control strategy, only the input and output are required, which makes the forging process easy to be controlled. In addition, because of the generalized ability and adaptability of neural networks, the proposed predictive controller can well deal with the time variance and nonlinearity of forging process. Two forging experiments demonstrate the feasibility and effectiveness of the proposed strategy. Moreover, comparing the proposed MPC strategy with the traditional MPC approach and PID controller, it can be found that the proposed MPC strategy is the most effective control approach for the practical forging process.

A deep belief network to predict the hot deformation behavior of a Ni-based superalloy

The hot deformation behavior of a Ni-based superalloy is studied by hot compressive experiments. The true stress is found to be highly affected by the deformation parameters, including strain rate and deformation temperature. The true stress dramatically decreases with decreasing strain rate or increasing deformation temperature. A deep belief network (DBN) model is developed for predicting true stress of the studied superalloy based on the experimental data. The structure of the developed DBN model is optimized layer by layer. The high accuracy indicates that the developed DBN model is able to effectively characterize the hot deformation behavior of the studied Ni-based superalloy. Moreover, the developed DBN model also has an excellent interpolation ability.