Nuclide energy spectrum generation method based on generative adversarial network

Background Obtaining nuclide energy spectra data poses significant challenges in practical applications, as traditional methods struggle to efficiently generate high-quality spectra. This paper proposes a novel approach based on a conditional generative adversarial network (CGAN) to address these limitations and facilitate the generation of accurate nuclide energy spectra.
Methods The proposed method utilizes a bidirectional long short-term memory (BiLSTM) network to capture temporal features of the data. In addition, Wasserstein distance with gradient penalty (WGAN-GP) is incorporated to optimize the training process, significantly enhancing the stability of the model.
Results Experimental results demonstrate that the generated nuclide spectra achieve remarkable performance in key evaluation metrics, including Fréchet inception distance (FID) and nuclide spectrum resolution, validating the effectiveness of the proposed approach.
Conclusions The proposed method provides an efficient and robust solution for generating nuclide energy spectra, offering significant potential for practical applications in nuclear science and related fields.