[Paper Review] Distributed Real-Time HVAC Control for Cost-Efficient Commercial Buildings under Smart Grid Environment

Liang Yu, Di Xie+3 more|arXiv (Cornell University)|Sep 3, 2017

Smart Grid Energy Management27 references3 citations

TL;DR

This paper proposes a distributed real-time HVAC control algorithm, CDRA, that minimizes long-term energy and thermal discomfort costs in commercial buildings under smart grid conditions. Using Lyapunov optimization with virtual queues for indoor temperatures, CDRA operates without predicting uncertain parameters like electricity prices or outdoor temperatures, achieving significant energy cost reduction with minimal comfort sacrifice.

ABSTRACT

In this paper, we investigate the problem of minimizing the long-term total cost (i.e., the sum of energy cost and thermal discomfort cost) associated with a Heating, Ventilation, and Air Conditioning (HVAC) system of a multizone commercial building under smart grid environment. To be specific, we first formulate a stochastic program to minimize the time average expected total cost with the consideration of uncertainties in electricity price, outdoor temperature, the most comfortable temperature level, and external thermal disturbance. Due to the existence of temporally and spatially coupled constraints as well as unknown information about the future system parameters, it is very challenging to solve the formulated problem. To this end, we propose a realtime HVAC control algorithm based on the framework of Lyapunov optimization techniques without the need to predict any system parameters and know their stochastic information. The key idea of the proposed algorithm is to construct and stabilize virtual queues associated with indoor temperatures of all zones. Moreover, we provide a distributed implementation of the proposed realtime algorithm with the aim of protecting user privacy and enhancing algorithmic scalability. Extensive simulation results based on real-world traces show that the proposed algorithm could reduce energy cost effectively with small sacrifice in thermal comfort.

Motivation & Objective

• Minimize the long-term total cost—comprising energy and thermal discomfort costs—in multizone commercial buildings under dynamic smart grid pricing.
• Address the challenge of temporally and spatially coupled constraints with unknown future system parameters such as electricity prices, outdoor temperatures, and occupant comfort preferences.
• Develop a real-time control algorithm that does not require prediction of future system parameters or knowledge of HVAC power demand.
• Ensure user privacy and algorithmic scalability through a distributed implementation of the control framework.
• Provide theoretical performance guarantees and feasibility analysis for the proposed algorithm under stochastic system uncertainties.

Proposed method

• Formulate a stochastic optimization problem to minimize the time-averaged expected total cost, incorporating uncertainties in electricity price, outdoor temperature, occupant comfort level, and external thermal disturbances.
• Apply Lyapunov optimization techniques by constructing virtual queues associated with indoor temperatures of each zone to stabilize system dynamics without requiring knowledge of HVAC power demand.
• Design a cost-aware distributed real-time algorithm (CDRA) that jointly minimizes the total cost and stabilizes the virtual queues using a drift-plus-penalty framework.
• Implement a distributed control architecture where each zone independently computes its control action based on local information, enhancing scalability and privacy.
• Use binary search to tune the penalty weight parameter V in the Lyapunov optimization framework to balance cost minimization and queue stability.
• Theoretical analysis proves that the algorithm achieves a time-averaged cost within O(1/V) of the optimal, with bounded queue stability and feasibility under i.i.d. system parameters.

Experimental results

Research questions

• RQ1 Can a real-time HVAC control algorithm be designed to minimize long-term total cost without predicting uncertain future parameters like electricity prices or outdoor temperatures?
• RQ2 How can virtual queues for indoor temperatures be used to replace direct knowledge of HVAC power demand in Lyapunov-based optimization?
• RQ3 What is the performance guarantee of a distributed, privacy-preserving HVAC control algorithm under stochastic system uncertainties?
• RQ4 To what extent can energy cost be reduced while maintaining acceptable thermal comfort levels in multizone commercial buildings?
• RQ5 How does the proposed algorithm compare to centralized or predictive approaches in terms of cost reduction and robustness under uncertainty?

Key findings

• The proposed CDRA algorithm reduces energy costs significantly with only a small increase in thermal discomfort, as validated by simulations using real-world data.
• The algorithm achieves a time-averaged total cost within O(1/V) of the optimal, with the bound becoming tighter as the penalty weight V increases.
• Theoretical analysis confirms that the algorithm maintains bounded queue stability and feasible operation under i.i.d. system parameters, ensuring long-term performance guarantees.
• The distributed implementation successfully protects user privacy and enhances scalability, as each zone operates independently based on local measurements.
• Simulation results show that CDRA effectively handles uncertainties in electricity prices, outdoor temperatures, and occupant comfort levels without requiring future predictions.
• The algorithm maintains indoor temperatures within acceptable bounds (T_i^min ≤ T_i,t ≤ T_i^max) across all zones and time slots, ensuring thermal comfort compliance.