摘要

High incidence and destructiveness of forest fire determine the importance of forest fire prediction or early detection. we find that the observation of local weather and human behaviors are the most correlated to forest fire important factors. In this paper, a fuzzy inference and big data analysis algorithm is proposed to assess fire risk and calculated the quantitative potential fire risk. These factors converted to triangular fuzzy numbers are calculated by fuzzification and output fire rating level. The continuous 24 hour weather information, which can reflect the high accurate status of forest environment, is collected by the rechargeable wireless sensor network. The fuzzy reasoning system was evaluated for Nanjing City region, the capital of Jiangsu Province, China. The high possibility of potential forest fires risk can be measured by this algorithm, therefore we need to pay more attention to prevent forest fire condition.