LDformer:a parallel neural network model for long-term power forecasting

Accurate long-term power forecasting is important in the decision-making operation of the power grid and power consumption management of customers to ensure the power system’s reliable power supply and the grid economy’s reliable operation.However,most time-series forecasting models do not perform well in dealing with long-time-series prediction tasks with a large amount of data.To address this challenge,we propose a parallel time-series prediction model called LDformer.First,we combine Informer with long short-term memory(LSTM)to obtain deep representation abilities in the time series.Then,we propose a parallel encoder module to improve the robustness of the model and combine convolutional layers with an attention mechanism to avoid value redundancy in the attention mechanism.Finally,we propose a probabilistic sparse(ProbSparse)self-attention mechanism combined with UniDrop to reduce the computational overhead and mitigate the risk of losing some key connections in the sequence.Experimental results on five datasets show that LDformer outperforms the state-of-the-art methods for most of the cases when handling the different long-time-series prediction tasks.

What kind of optical model potentials should be used for deuteron stripping reactions?

This paper presents the results of a study that compares CTOM, a microscopic optical model potential(OMP), which is an optical model co-created by the China Nuclear Data Center & Tuebingen University, to CH89, which is a typical phenomenological OMP.The respective OMPs were tested by applying them to the modelling of nucleon elastic scattering and(d,p) transfer reactions involving14C,36S, and58Ni targets at both low and relatively high energies. The results demonstrated that although both potentials successfully accounted for the angular distributions of both the elastic scattering and transfer cross sections, the absolute values of the transfer cross sections calculated using CTOM were approximately 25% larger than those calculated using CH89. This increased transfer cross sections allowed CTOM to produce single particle strength reduction factors for the three reactions that were consistent with those extracted from(e,e′p) reactions as well as with more recent(p,2p) and(p,pn) reactions. Notch tests suggested that nucleon elastic scattering and transfer reactions are sensitive to different regions of the OMP;accordingly,phenomenological OMPs, which are constrained only by elastic scattering cross sections, may not be sufficient for nucleon transfer reactions. Therefore, we suggest that microscopic OMPs, which reflect more theoretical considerations, should be preferred over phenomenological ones in calculations of direct nuclear reactions.