Uma abordagem para redução para a modelagem da consciência de disponibilidade em nós sensores de rede de sensores sem fio

• Ano de defesa: 2010
• Autor(a) principal: Adriano Borges da Cunha
• Tipo de documento: Tese
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Federal de Minas Gerais; UFMG
• Palavras-chave em Português: Modelos de bateria; Rede de sensores sem fio; Consciência de disponibilidade energética; Captura de energia do ambiente; Engenharia elétrica
• Link de acesso: http://hdl.handle.net/1843/COLT-8C7GX5

his work aims at modelling the Awareness of Available Energy (CODE in portuguese) in Wireless Sensor Network (WSN) nodes. This technique consists in providing the information of the amount of available energy in each sensor node at any time, making it possible to maintain, implement self-management techniques, plan quality of service (QoS) metrics and operational viability of WSN. In order to accomplish this it is necessary to model the energy supply unit of WSN nodes. In case of a WSN with no energy harvesting capability, this supply is based only on batteries. Thus, the implementation of CODE consists of the modelling of batteries. In case of a Energy-Harvesting WSN, the problem of estimating the amount of available energy consists of modelling and efficiently managing multiple sources of energy. This work is divided in two phases. The first phase is the elaboration, from a qualitative and quantitative approach, the first-generation of battery models for WSN, based on a book-keeping system method, that include at least the effects of discharge rate and relaxation. The second phase is the project and modelling of a simple and low cost energy supply unit with a solar energy harvest device, enabled with CODE and maximum-power-point tracking (MPPT). Experiments show that the global efficiency diverges from the optimal operating set point by less than 9%, and under certain conditions reaching 0%, confirming that it is also possible to adjust the maximum-power-point according to load demands. Characterization of an unit composed of a mini solar panel and supercapacitors, resulted in a new energy-efficient approach, that enables WSN nodes to obtain energy from three different sources: supercapacitors, back-up battery and a solar panel.