Developing semi-supervised variational autoencoder-generative adversarial network models to enhance quality prediction performance

One common serious issue of training a prediction model is that the process data significantly outnumber the quality data. Such discrepancy exists because of the time lag for obtaining quality data. This paper proposes semi-supervised variational autoencoder-generative adversarial network (S-2-VAE/GAN), that is able to make use of all the data even with some missing quality data. The key idea in S-2-VAE/GAN is the capability of enhancing the performance of the decoder/generator in learning the true distribution of both process and quality data in a competition between the decoder/generator and the discriminator in S-2-VAE/GAN through Nash Equilibrium, allowing the model to improve the qualifies of reconstruction and prediction data. The S-2-VAE/GAN model is also flexible enough to automatically adjust itself according to the input data. If the quality data are missing, the model can fill up the data through the trained prediction model and the same network structure defined in the supervised case can still be re-used. With the probabilistic distribution format, the proposed method is also capable of capturing the nonlinear feature of the process and representing the stochastic nature of operating plants. The results of the numerical case and the industrial case in this paper show that S-2-VAE/GAN outperforms conventional methods in terms of predictabilities of the missing quality data.

Automated summary

The S-2-VAE/GAN model enhances the performance of the decoder/generator in learning the true distribution of process and quality data through a competition with the discriminator, improving predictabilities of the missing quality data. The model is also flexible in adjusting to input data and filling missing quality data, ultimately capturing nonlinear features and representing stochastic nature effectively.