In the mineral processing industry, accurately predicting concentrate grade can help engineers adjust process parameters in advance and improve flotation performance. However, the prediction accuracy of concentrate grade has been restricted by small sample sizes, high-dimensional data, and complex temporal correlations in actual mineral processing. To address the predication challenges associated with small sample data, a time-series data generation model called LS-WGAN is proposed, which combines the Wasserstein generative adversarial network (Wasserstein GAN) and long short-term memory (LSTM) neural network. The LSTM network is mainly used to capture the time correlation in mineral processing data, while the Wasserstein GAN generates samples similar to the original data distribution for data augmentation. To improve the prediction accuracy of the concentrate grade, a mineral processing prediction model called C-LSTM is established. The prediction accuracy of the proposed method is verified through experiments based on real froth flotation process data.