Time Series Forecasting on Engineering Systems Using Recurrent Neural Networks.

Modern large scale processing and manufacturing systems cover a wide array and large number of assets that need to work together to ensure that plant is generating output reliably and at the desired yield rate, such as the viscosity of quench oil in styrene cracking system. However, due to the complexity of the overall process, it is important to consider the entire plant as a network to identify deterioration patterns and forecast condition.