Model Predictive Control of a Geothermally Heated Bridge Deck

This paper describes the application of model predictive control (MPC) to a geothermal bridge deck heating system. The control system integrates concepts of MPC with a first-principles bridge deck model and hourly computerized national weather service (NWS) forecasts to prevent bridge icing without the use of salt or other chemical de-icing materials. The proactive nature of the control system maximizes motorists' safety and bridge life while minimizing system operating costs. Results are presented for a snow event on a %-scale, hydronically-heated bridge deck. The effect of MPC error penalty and move suppression weighting on bridge deck heating performance is discussed.