Multi-agent control system with intelligent optimisation for building energy and comfort management

Abstract

Smart, sustainable and energy-efficient buildings have recently become a trend for intelligent and green building industry. One of the challenges for such a building is to minimise power consumption without compromising the occupant comfort. It is possible to sustain high comfort level with minimal energy consumption through intelligent multi-agent control and optimisation. Our work focuses on a new algorithmic aspect that makes buildings more intelligent, resource efficient and comfortable for occupants. A multi-agent control system with intelligent optimisation is proposed to optimise heating, ventilation and airconditioning (HVAC) processes using computational intelligence approaches. The optimum solutions generated through the optimisation engine and task scheduling for each agent will be used to sustain a high level of occupant comfort with minimal power consumption, increase energy utilisation efficiency, and consequently reduce energy cost. Multi-agent control system with intelligent optimisation 61 Experimental results and comparisons with other evolutionary approaches demonstrate the overall performance and potential benefits of the new framework.

RETRACTED: Hybrid multi agent optimization for optimal battery storage using micro grid

“…The MAS concept is just a decentralized database made up of a number of autonomous algorithms which have been developed to improve efficiency and spread burdens. Loading agency, production entity, centralized operative, and storage intermediary are the four types of MAS elements (Kan et al, 2016; Liu et al, 2018; Mets et al, 2012; Sharma et al, 2018). The workload and storage agents collaborated with the generating agent, splitting the biggest issue into successive smaller problems to resolve complex goals.…”

An Embedded Platform for Testbed Implementation of Multi-Agent System in Building Energy Management System

“…The main objectives of the energy management systems [9][10][11][12][13][14][15][16][17][18][19][20][21][22][23][24][25][26][27] are usually to minimize the energy cost and/or maximize user comfort. In addition to these two optimization objectives, the objective function of maximizing the energy usage from the local renewable energy source is also employed [28,29].…”

“…The MAS-based BEMS could be classified into three groups, i.e., based-on: the number of criteria to be optimized, the optimization algorithm, and the implementation platform. In the first group, they are divided into single-objective optimization [10][11][12][13][14][15][16][17] and multi-objectives optimization [18][19][20][21][22][23]. Based on the optimization algorithm, they are divided into the conventional-based optimization techniques [10,11,13,15,17,21,23] and the artificial intelligent (AI)-based optimization techniques [12,14,16,[18][19][20]22].…”

“…Furthermore, our proposed optimization technique, which is used to minimize the energy cost while maximizing the user comfort, utilizes a few parameters for calculating the objective function. Instead of using both the energy cost and the comfort parameters in the objective function explicitly [18][19][20][21][22][23]28,29], our method uses the energy cost parameter only, since the comfort parameters could be represented in the term of energy cost parameter as described in the following. The power consumption in each room, which is calculated by the load agent, reflects the user comfort (the thermal comfort and the illumination comfort) of the room, in the sense that high power consumption represents high comfort, and low power consumption represents low comfort.…”

“…A system to predict the energy demand of the building using the Artificial Neural Network (ANN) is proposed in [5]. The ANN model is trained using the dataset of monthly historical energy consumption of the building.Due to the distributed components (sensors, actuators, generators, loads) of the HEMS/BEMS, an intelligent multi-agent system (MAS) is widely adopted [6][7][8][9][10][11][12][13][14][15][16][17][18][19][20][21][22][23]. The MAS is a distributed control system where each agent works autonomously and coordinates with each other to achieve the global goal.…”