Time-series prediction is one of the major methodologies used for fault prediction. The methods based on recurrent neural networks have been widely used in time-series prediction for their remarkable non-liner mapping...Time-series prediction is one of the major methodologies used for fault prediction. The methods based on recurrent neural networks have been widely used in time-series prediction for their remarkable non-liner mapping ability. As a new recurrent neural network, reservoir neural network can effectively process the time-series prediction. However, the ill-posedness problem of reservoir neural networks has seriously restricted the generalization performance. In this paper, a fault prediction algorithm based on time-series is proposed using improved reservoir neural networks. The basic idea is taking structure risk into consideration, that is, the cost function involves not only the experience risk factor but also the structure risk factor. Thus a regulation coefficient is introduced to calculate the output weight of the reservoir neural network. As a result, the amplitude of output weight is effectively controlled and the ill-posedness problem is solved. Because the training speed of ordinary reservoir networks is naturally fast, the improved reservoir networks for time-series prediction are good in speed and generalization ability. Experiments on Mackey–Glass and sunspot time series prediction prove the effectiveness of the algorithm. The proposed algorithm is applied to TE process fault prediction. We first forecast some timeseries obtained from TE and then predict the fault type adopting the static reservoirs with the predicted data.The final prediction correct rate reaches 81%.展开更多
By using an instrumented impact pendulum, the force versus time curves of 7075-T651 aluminum welds were obtained from standard Charpy-V samples. Considering the force-time curves and constant impact velocity, the frac...By using an instrumented impact pendulum, the force versus time curves of 7075-T651 aluminum welds were obtained from standard Charpy-V samples. Considering the force-time curves and constant impact velocity, the fracture energies for different zones were quantified. A fracture energy improvement for the HAZ(33.6 J) was observed in comparison with the weld metal(7.88 J), and base metal(5.37 J and 7.37 J for longitudinal and transverse directions, respectively). This toughness increment was attributed to the microstructural transformation caused by the thermodynamic instability of η′ precipitates during the welding. Fracture energy for weld metal was higher than that for base metal, probably due to pores created during solidification. Regarding the dynamic yielding force obtained from the force-time curves, an approximation to the dynamic yield strength for weld, HAZ and base metal was determined. Fracture surfaces revealed an intergranular failure for base metal in longitudinal direction, whereas a predominately brittle failure(cleavage) with some insights of ductile characteristics was observed for the transverse direction. In contrast, a ductile failure was observed for weld metal and HAZ.展开更多
基金Supported by the National Natural Science Foundation of China(61074153)
文摘Time-series prediction is one of the major methodologies used for fault prediction. The methods based on recurrent neural networks have been widely used in time-series prediction for their remarkable non-liner mapping ability. As a new recurrent neural network, reservoir neural network can effectively process the time-series prediction. However, the ill-posedness problem of reservoir neural networks has seriously restricted the generalization performance. In this paper, a fault prediction algorithm based on time-series is proposed using improved reservoir neural networks. The basic idea is taking structure risk into consideration, that is, the cost function involves not only the experience risk factor but also the structure risk factor. Thus a regulation coefficient is introduced to calculate the output weight of the reservoir neural network. As a result, the amplitude of output weight is effectively controlled and the ill-posedness problem is solved. Because the training speed of ordinary reservoir networks is naturally fast, the improved reservoir networks for time-series prediction are good in speed and generalization ability. Experiments on Mackey–Glass and sunspot time series prediction prove the effectiveness of the algorithm. The proposed algorithm is applied to TE process fault prediction. We first forecast some timeseries obtained from TE and then predict the fault type adopting the static reservoirs with the predicted data.The final prediction correct rate reaches 81%.
基金CONACy T (project CB 177834)SIP-IPN for the funds given to conduct this research
文摘By using an instrumented impact pendulum, the force versus time curves of 7075-T651 aluminum welds were obtained from standard Charpy-V samples. Considering the force-time curves and constant impact velocity, the fracture energies for different zones were quantified. A fracture energy improvement for the HAZ(33.6 J) was observed in comparison with the weld metal(7.88 J), and base metal(5.37 J and 7.37 J for longitudinal and transverse directions, respectively). This toughness increment was attributed to the microstructural transformation caused by the thermodynamic instability of η′ precipitates during the welding. Fracture energy for weld metal was higher than that for base metal, probably due to pores created during solidification. Regarding the dynamic yielding force obtained from the force-time curves, an approximation to the dynamic yield strength for weld, HAZ and base metal was determined. Fracture surfaces revealed an intergranular failure for base metal in longitudinal direction, whereas a predominately brittle failure(cleavage) with some insights of ductile characteristics was observed for the transverse direction. In contrast, a ductile failure was observed for weld metal and HAZ.
