A Multi-Condition Sequential Network Ensemble for Industrial Energy Storage Prediction Considering the Condition Switching Characteristics

As a crucial storage and buffering apparatus for balancing the production and consumption of byproduct gases in industrial processes, accurate prediction of gas tank levels is essential for optimizing energy system scheduling. Considering that the continuous switching of the pressure and valve status (mechanism knowledge) would bring about multiple working conditions of the equipment, a multi-condition time sequential network ensembled method is proposed. In order to especially consider the time dependence of different conditions, a central-wise condition sequential network is developed, where the network branches are specially designed based on the condition switching sequences. A branch combination transfer learning strategy is developed to tackle the sample imbalance problem of different condition data. Since the condition or status data are real-time information that cannot be recognized during the prediction process, a pre-trained and ensemble learning approach is further proposed to fuse the outputs of the multi-condition networks and realize a transient-state involved prediction. The performance of the proposed method is validated on practical energy data coming from a domestic steel plant, comparing with the state-of-the-art algorithms. The results show that the proposed method can maintain a high prediction accuracy under different condition switching cases, which would provide effective guidance for the optimal scheduling of the industrial energy systems.

I. Introduction

STEEL industry generally accompanies with high energy consumption and emission. The main reason comes from the imbalance of energy generation and consumption. As an important secondary energy, the byproduct gases are recovered and consumed in steel-making processes. An effective usage of byproduct gases can not only save economic costs, but also reduce the carbon emissions of enterprises [1]. While the gas tank levels in byproduct gas systems are important monitoring indexes for evaluating the energy balance, and formulating reasonable energy scheduling policies [2].