Taiwan has suffered from widespread haze and poor air quality during recent years, and the control of indoor air quality has become an important topic. This study relies on Multi-Agent theory in which collected air quality was used in calculations and after agents make decisions in accordance with pre-written rules to construct and indoor air quality control system and conflict resolution mechanism, which will serve to maintain a healthy and comfortable indoor environment. As for implementation, the simulated system used the Arduino open source microcontroller system to collect air quality data and turn on building equipment in order to improve indoor air quality. This study also used the graphic control program LabVIEW to write a control program and user interface. The implementation verifies the feasibility of applying multi-agent theory to air quality control systems, and an Individual intelligent agent has the basic ability to resolve their own conflicts autonomously. However, when there are multiple factors and user status are simultaneously involved in the decision-making, it is difficult for the system to exhaust all conflict conditions, and when context control surpassing the restrictions of binary logic rule-based reasoning, it is necessary to change the algorithm and redesign the system.
Indoor Air Quality Control Using Backpropagated Neural Networks
