Observação e modelagem dos processos físicos envolvidos naformação das descargas atmosféricas

• Ano de defesa: 2017
• Autor(a) principal: Miguel de Brito Guimaraes Neto
• Tipo de documento: Tese
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Federal de Minas Gerais; UFMG
• Palavras-chave em Português: Engenharia elétrica; Raio
• Link de acesso: http://hdl.handle.net/1843/BUOS-ATELG3

This work addressed features of positive and negative leaders in the phase just preceding the attachment in negative CG lightning, based on experimental results. These results consist of simultaneous records of current, electric fields, luminosity and high-speed videos obtainedat Morro do Cachimbo Station (MCS) and presented for the first time in literature. The developments involved the improvement of the local measuring systems to enhance the reliability of their results and the monitoring of the related quantities during two storm seasons. The interpretation of the results led to significant contributions in three fields. First, regarding the propagation of negative leaders, the results revealed a relatively small variation of the leader speed in the last few hundred meters propagated distance in all measured events. Also, a rareobservation was documented, consisting of the continuous propagation of the negative downward leader preceding a first stroke, different from the usual stepwise propagation. With respect to upward positive leaders, the work identified a common initiation condition in all currents measured recently at MCS, consisting of a 4-A threshold continuous current measured at the tower base. This threshold corresponds to an electric field at ground, measured 50 m away from the tower, of about 60 kV/m. Based on experimental evidences and the analysis of the literature, the work concluded that positive leaders develop continuously, without stepping. The propagation speed of the recorded positive leaders exhibited high dispersion, around an average value significantly lower than that of the negative leaders. Finally, for the first time, the work determined pairs of striking distance (SD) and directly measured current of first return strokes. The SDs were determined according to four different methodologies. It was shown that SD determined solely from high-speed video can be significantly underestimated. The use of the 4-A condition for the upward leader initiation was introduced in SD calculation and a new methodology to determine the SD solely from records of current was developed. This methodology was applied to determine the SD of 17 first return strokes. The results confirmedthe trend of striking-distance increase with increasing peak current and denoted a very high dispersion of the striking distance in events exhibiting comparable peak currents, attributed to unpredictable geometrical distribution of the multiple branches in the individual negativeleaders.