A Markov Process Approach to Ensemble Control of Smart Buildings

Abstract

This paper describes a step-by-step procedure that converts a physical model of a building into a Markov Process that characterizes energy consumption of this and other similar buildings. Relative to existing thermo-physics-based building models, the proposed procedure reduces model complexity and depends on fewer parameters, while also maintaining accuracy and feasibility sufficient for system-level analyses. Furthermore, the proposed Markov Process approach makes it possible to leverage real-time data streams available from intelligent data acquisition systems, which are readily available in smart buildings, and merge it with physics-based and statistical models. Construction of the Markov Process naturally leads to a Markov Decision Process formulation, which describes optimal probabilistic control of a collection of similar buildings. The approach is illustrated using validated building data from Belgium.

Privacy-Aware Load Ensemble Control: A Linearly-Solvable MDP Approach

“…The load ensemble can be represented by a discrete-time, discrete-space MDP similar to our prior work in [9]- [12], [15], [16]. We consider a LS-MDP with state space A, time space T , default transition probability matrix P ∈ R n×n , n = |A|, controlled transition probability matrix P t ∈ R n×n , n = |A| and utility of the aggregator is obtained by discretizing controllable parameters of the loads in the ensemble (e.g., power consumption or temperature levels) and time space T = {t, t + 1, • • • } represents discrete time intervals (e.g., hours or minutes).…”

A Hierarchical Approach to Multienergy Demand Response: From Electricity to Multienergy Applications

“…Using the three-step procedure described in Section IV-A and illustrated in Fig. 5, we construct the MP for a portfolio of residential households in Belgium, where each household is an 'average' low-energy building, in which the day and night zones have a surface area of 132 m 2 and 138 m 2 , respectively, [79]. In this built environment, individual heating systems consist of an air-coupled heat pump and a back-up electric resistance heater, which supply heat to the floor heating system in the day and night zones, i.e.…”

“…The MDP optimization formulated in (2)-(4) is a LS-MDP as introduced in [80]- [82] and discussed in [78], [79], [83], [84]. LS-MDP problems can be solved analytically and the 2 Note that although P αβ is defined as time-independent, one can model it as time-dependent if there is enough observation data to construct a multiperiod MP.…”

“…5 and includes the following three steps: (i) building Figure 5. The proposed three-step procedure to construct MP for optimal ensemble control of buildings, [79].…”

“…The proposed three-step procedure to construct MP for optimal ensemble control of buildings,[79].data generation and aggregation, (ii) state-space definition and reduction, and (iii) construction and validation of the resulting MP. These steps are further detailed below.…”