Anastasios I. Dounis

Design of a fuzzy set environment comfort system

This paper presents the design of a fuzzy reasoning expert system for the achievement of thermal and visual comfort in buildings. This system does not demand the precise mathematical model of the building to achieve the control law but uses high-level control variables such as thermal and visual comfort. The powerful interactions of the passive components and of the comfort subjectivity match with the application of the fuzzy control theory entirely. Mathematical models are presented, where the actions of the actuators are applied. The design of the rule base is described and, finally, the system is evaluated by using extensive, worst-case, simulation results.

Design of a fuzzy system for living space thermal-comfort regulation

The present paper describes the design of a living space comfort regulator using fuzzy logic. Comfort is a fuzzy concept, different for different people and depending on the work done in the space. The paper describes the structure of the system, the available measurements and the available actuators, the measurement fuzzification process and the defuzzification method. Particular attention is paid to the proper selection of the rules in the knowledge base and the design of the inference engine. Finally the system is tested, and shows satisfactory performance. General design guidelines are given, including the case of spaces having different actuators.

Building visual comfort control with fuzzy reasoning

Abstract Approximate reasoning is, in many cases, a more successful control strategy than a classically designed control scheme. Human reasoning can be reasonably well modelled by fuzzy logic. In this paper, fuzzy logic is used to develop a control scheme for visual comfort in buildings used either for home or office. The fuzzy controller is developed, and a linguistic type of control algorithm is presented. Visual comfort and the relevant processes are described. Mathematical models are used to calculate lighting and glare in the building.

Comparison of Conventional and Fuzzy Control of Indoor Air Quality in Buildings

This article compares various techniques for the control of indoor air quality in a naturally ventilated building: ON-OFF, proportional-integral-derivative PID, proportional-integral with deadband PIdb, and fuzzy control. The control system for the fuzzy controllers is a typical two input carbon dioxide concentration and its time derivative, one output change in window opening area system. Two different fuzzy controllers are modeled: one based on Mamdanis max-min inference, and the other using Guptas multivariable structure. The results show that the fuzzy controllers and the Pldb controller are much more suitable than the PID or the ON-OFF controllers. The number of cycles of the controlled parameter i.e., the window opening fluctuations is actually much smaller. Minor differences between the two fuzzy controllers are observed.

Artificial intelligence for energy conservation in buildings

Abstract The problem of energy conservation in buildings is a multidimensional one. Researchers from a variety of disciplines have been working on this problem. It remains a challenging and yet rewarding study. In the past three decades, a plethora of scientific and technological publications on energy conservation in buildings have been presented in international journals. In this work, we discuss the potentiality of artificial intelligence (AI) as a design tool in building an automation system. The application of contemporary AI techniques creates intelligent buildings with the following main goals: energy efficiency, comfort, health and productivity in living spaces. Two modern domains of AI that are widely used in buildings are computational intelligence (CI) or soft computing and distributed artificial intelligence (DAI). DAI includes intelligent agents (IAs), multi-agent systems (MASs) and ambient intelligence. However, there is a lack of systematic review of research efforts and achievements mainly on IA and MAS domains. This chapter briefly presents expert systems and CI techniques and outlines how they operate. The major objective of this chapter is to illustrate how IAs and MASs may play an important role in conserving energy in buildings.

A fuzzy cognitive maps-petri nets energy management system for autonomous polygeneration microgrids

Autonomous polygeneration microgrids (APM) are a relatively new approach in covering specific needs like power, potable water and fuel for transportation, in remote areas. This approach has been proved to be technically feasible nowadays and even present itself as an economically viable investment. The initial management system built for this approach is a simple ON/OFF supervisor which can make the APM operate, but not in an optimal way. The devices cannot be operated in part load and as a consequence there is little room for optimization. A combined fuzzy cognitive maps (FCMs)-petri nets (PN) approach has been developed for the energy management of such a system. The PN is used as an activator in the fuzzy cognitive map structure so as to enable different FCMs to be activated depending on the state of the microgrid. This combination forms an integrated approach to the energy management of the microgrid. Using this approach considerable optimization in the design and operation of the microgrid is possible. A methodology for simultaneous and interactive optimization of the energy management system along with the sizing of the various devices of the actual microgrid is implemented. A software platform consisting of TRNSYS, TRNOPT and GenOPT software packages was used for simulation and optimization. Particle swarm optimization is applied both for the sizing of the system and the optimization of the FCM weights and PN parameters. Two microgrids were designed, one based on the FCM-PN energy management system (FPEMS) and one on the ON/OFF approach. The results show that FPEMS manages the energy flows more effectively throughout the year which leads to a considerable decrease in the sizing of the various components of the microgrid.

Indoor air-quality control by a fuzzy-reasoning machine in naturally ventilated buildings

This paper investigates the performance of a fuzzy reasoning machine for the control of indoor air quality in naturally ventilated buildings. Simulations have been performed using a new airflow and pollutant transport model, which has been developed and validated for this purpose; CO2 concentration was used as the indoor-air quality (IAQ) index for these simulations. Results have shown that satisfactory IAQ levels can be maintained, while good stability of the control parameter (i.e. window opening area) was achieved. The impact of such a controller on indoor-air temperature was also studied. The performances were not as good as expected, but were not negligible when compared with the normal conditions of use of the building.

Customer-centered control system for intelligent and green building with heuristic optimization

The big challenge of the control system for the intelligent and green building is to maintain the maximum customer comfort with minimum energy consumption. In the paper, a customer-centered multi-agent control system is proposed to meet the challenge. Four different kinds of agents are described, which are switch agent, central coordinator-agent, local controller-agents and load agent. Particle swarm optimization (PSO) is utilized to optimize the overall system and to enhance the intelligence of the system. The preferences of customers are included in the control system design. Customers are offered the flexibility to define control parameters through a graphical user interface (GUI) according to their own preferences.

Intelligent Coordinator of Fuzzy Controller-Agents for Indoor Environment Control in Buildings Using 3-D Fuzzy Comfort Set

In this paper, we develop an intelligent coordinator (IC) of fuzzy controller-agents (FCAs) for indoor environmental conditions control in buildings using a 3-D fuzzy comfort concept as an information granule. The proposed intelligent coordination model has hierarchical structure. This centralized coordinator consists of two subsystems the master and slave agents. These subsystems are implemented by fuzzy logic rules. The master agent evaluates the energy efficiency of the building and comfort and the fuzzy inference mechanism produces signals that activate slave agent and change the set points of the controllers. The slave agent is a fuzzy negotiation machine (FNM), which compensates the interaction of the FCAs and manages to avoid conflicts between them. The FCAs are activated when some conditions determined by the slave agent are satisfied, otherwise they stay inactive. Finally, the applicability of the suggested system is demonstrated via TRNSYS-MATLAB computer simulation.

IMPLEMENTATION OF ARTIFICIAL INTELLIGENCE TECHNIQUES IN THERMAL COMFORT CONTROL FOR PASSIVE SOLAR BUILDINGS

Artificial Intelligence techniques are used to control thermal comfort levels in a passive solar building. The controller, as well as the necessary group of rules, are described and analysed. Fuzzy logic is used for the first time in passive building control.