A Energy Saving Control Strategy For Heating, Ventilation, Air-Conditioner Systems

The rapid development of science and technology in society today, people of their quality of life demands has been further improved. The combination of HVAC and building can provide people with a more comfortable living environment and office environment. However, the problem that the arising of energy consumption of HVAC can not be ignored and avoided, especially in the relatively scarce resource today. The energy that HVAC consumes is mainly non-renewable and high quality. Therefore, the study of energy-efficient of HVAC has a very important practical significance.In the energy saving of HVAC, the detecting of temperature is a very important aspect. In this paper, we use the wireless sensor networks to get the temperature of the room. The reason why we choose the wireless sensor network is the deployment process of the wireless sensor network does not require make changes on existing wire-line. Otherwise, it has also been favored by many scholars because its low cost, ease of deployment and so on. This network has a good prospect.Energy consumptions and system delays are two critical factors to be considered when designing wireless sensor networks. In the first two chapters, we developed an indoor temperature monitoring system with wireless sensor networks, including designs of nodes, network topology and upper monitoring software. In contrast to existing indoor temperature monitoring system, we adopted the sensed temperature as a variable of the system. By dynamically adjusting the sleep/wake duty-cycle of both sensor nodes and relay nodes, in case a temperature anomaly occurs, the system delay is relatively short so that an alarm can be triggered in time. In contrast, if the sensed temperature is normal, the system can ensure a good energy efficiency performance. The sensor nodes wakes up and sense the surrounding temperature periodically using sleep/wake duty cycles. Otherwise, the sensor nodes send the data to the destination node via multi-hop relaying nodes in an anycast way. The destination node receives the packets from the relaying nodes and transfers every packet to PC through serial communication, where the measured temperature is extracted and displayed, and triggers an alarm if the sensed temperature gets closer to the pre-fixed alarm threshold. Through an implementation using TelosB with TinyOS, we built the system and analyzed the system delay performance both theoretically and practically. We also gave an analysis about other performances like energy consumptions. We can show that the system runs as expected and have better performances than exiting systems.With the development and promotion of smart grid, the research of energy-saving of HVAC and temperature control in the smart grid also has a very important practical and long-term significance. The smart grid is the new generation of electricity grid. It's a new kind of technology which has newly emerged in North America and developed very fast. Compared with electricity grid which is labeled with high construction cost, high energy cost, smart grid can more efficiently collect information from all areas and use different kinds of energy including renewable sources like solar energy, wind energy, and provide the new policy of time-varying electricity price. In the third chapter, we propose an optimal dynamic algorithm to control the Heating, Ventilation, Air-Conditioner (HVAC) system in an indoor environment such as data center, a residential house or a factory that may generate a lot of heat. The time varying electricity price provides an opportunity for us to cut the electricity price and save energy. The problem we formulate here is to minimize the total energy cost of the HVAC system while at the same time keep the indoor temperature within a prefixed range. The methodology we use is Lyapunov optimization and we deal with the hard temperature constrains by introducing a penalty function. The algorithm requires no future knowledge of power consumption rate of the appliances and the electricity process. Finally we proved the efficiency of our algorithm in Matlab.