Development of reduced order thermal dynamic models for building load flexibility of an electrically-heated high temperature thermal storage device

Abstract

A control-oriented model of an Electrically heated Thermal Energy storage device (ETS) is presented. The ETS consists of bricks heated up to 871 °C with electric coils; heat is discharged with an airflow passing through channels in the bricks. The device stores energy during off-peak periods to meet future on-peak building heating loads. While its rationale is understood, improvement regarding operation strategies is needed. The charging/discharging could be adjusted according to predicted heating loads to reduce bills, and enhance building energy flexibility in its interaction with the grid. The model, to facilitate implementation of model predictive control (MPC), is based on the following features: (a) calibration of grey-box resistance-capacitance models with a heat exchanger model; (b) periodic state updates from sensors; (c) “effective” brick conductivity. A 1-capacitance model with state updates and “effective” brick conductivity was comparable to a detailed 140-capacitance model. Therefore, a simple model with modifications gives adequate predictions and thus can aid with MPC. Finally, strategies are evaluated to study energy flexibility potential. Using the ETS during the morning critical demand hours and anticipating the rise in setpoint temperature from nighttime setback, peak reduction 11% is obtained. A reduction of 73% is found when HVAC heating coil is limited.

Acknowledgments

The authors would like to thank researchers at Hydro-Québec LTE for their helpful collaboration and insight. Thank you also to the contributions of internal reviewers at Hydro-Québec and CanmetENERGY-Varennes.

Additional information

Funding

The financial support of the NSERC-Hydro-Québec Industrial Research Chair and the research project EIP-EU-BE-11 under the Canadian Federal Program of Energy Research and Development (PERD) is gratefully acknowledged. FRQNT (Fonds de recherche Nature et technologies) in Quebec and ASHRAE Graduate Student Grant-in-Aid is gratefully thanked for partial funding of PhD candidate Jennifer Date.