Energy modeling of thermal energy storage (TES) using intelligent stream processing system

Abstract

Thermal energy storage (TES) is the core element of renewable energy system (RES) and can considerably affect its overall efficiency. An effective thermal energy storage (TES) should enhance the stratification by restricting inlet mixing. In this paper, an experimental study is presented to evaluate the performance of thermal energy storage (TES). Discharging of the tank was conducted with different inlet flow rates to assess the effect of inlet mixing on thermal stratification. Results are quantified in terms of temperature distribution, MIX, and Richardson number and were visualized to predict the behavior of TES. In addition, the data parsing is done in live mode with ad-hoc built stream-processing data layer. Finally a methodology for time series prediction in the context of TES using high end LSTM network is framed. It was concluded that discharging rate of 800 l/h has the maximum mixing and thus the worst stratification, while prediction efficiency fell well within 5.2% of the error range.

Keywords

• Thermal energy storage (TES)
• Data streaming with RaspberryPi and Python
• Renewable Energy