Share

Formulation and application of an economic model predictive control scheme for connected thermostats.

Number: pap. 3194

Author(s) : ELLIS M. J., ALANQAR A.

Summary

An economic model predictive control (MPC) scheme, which is an MPC scheme equipped with an economicallyoriented objective function, is developed for a connected thermostat application. The economic MPC selects a setpoint from an occupant defined comfort range to minimize the HVAC power cost. Specifically, under a time-varying electric rate structure (e.g., time-of-use or real-time pricing), the economic MPC leverages the building mass as thermal energy storage to shift HVAC power consumption from high to low cost periods. The resulting economic MPC system includes a parameterized building thermal zone model, a parameter estimation procedure to identify the model parameters for a specific zone application, a state/disturbance estimator, a heat load disturbance forecaster, and an underlying optimal control problem formulation. Each of these features are tailored for broad application as a supervisory controller manipulating the zone temperature setpoint for a zone controlled by a thermostat. Given a lack of measurements available to estimate/measure HVAC power or load in a typical thermostat, the HVAC load is approximated via a filtered version of the thermostat equipment stage commands, which provides a normalized time-averaged approximation of the HVAC load. Simulation results are presented to demonstrate the effectiveness of the strategy.

Available documents

Format PDF

Pages: 10

Available

  • Public price

    20 €

  • Member price*

    15 €

* Best rate depending on membership category (see the detailed benefits of individual and corporate memberships).

Details

  • Original title: Formulation and application of an economic model predictive control scheme for connected thermostats.
  • Record ID : 30024744
  • Languages: English
  • Subject: Environment
  • Source: 2018 Purdue Conferences. 5th International High Performance Buildings Conference at Purdue.
  • Publication date: 2018/07/09

Links


See other articles from the proceedings (88)
See the conference proceedings

Indexing