Building occupancy simulation and data assimilation using a graph-based agent-oriented model

Authored by Xiaolin Hu, Sanish Rai

Date Published: 2018

DOI: 10.1016/j.physa.2018.02.051

Sponsors: No sponsors listed

Platforms: No platforms listed

Model Documentation: Other Narrative Pseudocode

Model Code URLs: Model code not found

Abstract

Building occupancy simulation and estimation simulates the dynamics of occupants and estimates their real-time spatial distribution in a building. It requires a simulation model and an algorithm for data assimilation that assimilates real-time sensor data into the simulation model. Existing building occupancy simulation models include agent-based models and graph-based models. The agent-based models suffer high computation cost for simulating large numbers of occupants, and graph-based models overlook the heterogeneity and detailed behaviors of individuals. Recognizing the limitations of existing models, this paper presents a new graph-based agent-oriented model which can efficiently simulate large numbers of occupants in various kinds of building structures. To support real-time occupancy dynamics estimation, a data assimilation framework based on Sequential Monte Carlo Methods is also developed and applied to the graph-based agent-oriented model to assimilate real-time sensor data. Experimental results show the effectiveness of the developed model and the data assimilation framework. The major contributions of this work are to provide an efficient model for building occupancy simulation that can accommodate large numbers of occupants and an effective data assimilation framework that can provide real-time estimations of building occupancy from sensor data. (C) 2018 Elsevier B.V. All rights reserved.

Tags

data assimilation Agent-based simulation Building occupancy simulation Graph-based modeling Sequential monte carlo methods Particle filters