No abstract

The outputs of the Chinese Academy of Sciences (CAS) Flexible Global Ocean-Atmosphere-Land System (FGOALS-f3-L) model for the baseline experiment of the Atmospheric Model Intercomparison Project simulation in the Diagnostic, Evaluation and Characterization of Klima common experiments of phase 6 of the Coupled Model Intercomparison Project (CMIP6) are described in this paper. The CAS FGOALS-f3-L model, experiment settings, and outputs are all given. In total, there are three ensemble experiments over the period 1979-2014, which are performed with different initial states. The model outputs contain a total of 37 variables and include the required three-hourly mean, six-hourly transient, daily and monthly mean datasets. The baseline performances of the model are validated at different time scales. The preliminary evaluation suggests that the CAS FGOALS-f3-L model can capture the basic patterns of atmospheric circulation and precipitation well, including the propagation of the Madden-Julian Oscillation, activities of tropical cyclones, and the characterization of extreme precipitation. These datasets contribute to the benchmark of current model behaviors for the desired continuity of CMIP.

No abstract

No abstract

Time series anomaly detection is widely used to monitor the equipment sates through the data collected in the form of time series. At present, the deep learning method based on generative adversarial networks (GAN) has emerged for time series anomaly detection. However, this method needs to find the best mapping from real-time space to the latent space at the anomaly detection stage, which brings new errors and takes a long time. In this paper, we propose a long short-term memory-based variational autoencoder generation adversarial networks (LSTM-based VAE-GAN) method for time series anomaly detection, which effectively solves the above problems. Our method jointly trains the encoder, the generator and the discriminator to take advantage of the mapping ability of the encoder and the discrimination ability of the discriminator simultaneously. The long short-term memory (LSTM) networks are used as the encoder, the generator and the discriminator. At the anomaly detection stage, anomalies are detected based on reconstruction difference and discrimination results. Experimental results show that the proposed method can quickly and accurately detect anomalies.