Data Augmentation for Intelligent Manufacturing with Generative Adversarial Framework

Yanxia Wang, Kang Li, Shaojun Gan, Che Cameron, Min Zheng
2019 2019 1st International Conference on Industrial Artificial Intelligence (IAI)  
Motivated by the global economy greatly shaped by the manufacturing technology, more research on the intelligent manufacturing is studied. This paper firstly introduces an energy monitoring and data acquisition system namely the Point Energy Technology, which has been developed by the team and installed in a local bakery. While there is always lack of data because of various reasons, such as measurement or transmission mistakes during data collection. To solve this problem, we introduce a
more » ... tive adversarial framework which is based on a game theory for data augmentation. This framework consists of two multi-layer perceptron networks-generator and discriminator. The upgrade framework with Q-net that extracts the latent variables from real data is proposed. To control the number of parameters, Q-net shares the structure with discriminator except the last layer. In addition, the two optimization methods, mini-batch gradient descent and adaptive moment estimation are adopted to tune the parameters. To evaluate the performance of these algorithms, the collected data from baking process is used in the experiment. Considering the reality, the missing data is processed into the state of missing completely at random (non-time series missing data). The experimental results illustrate that the latent generative adversarial framework with adaptive moment estimation could generated samples of good quality for non-time series missing data.
doi:10.1109/iciai.2019.8850773 fatcat:jsjnkhbzszgodneookx7mundxu