CAJ | 학술논문

本研究基于1953-2007年登陆或对浙江省有重大影响的台风历史案例数据，考虑影响台风灾害损失大小的主要因素有台风致灾因子、承灾体暴露性影响因子与承灾体脆弱性影响因子，运用主成分分析法对表示承灾体暴露性影响因素与承灾体脆弱性影响因素进行数据处理，提取主成分作为RBF神经网络模型的输入，从而建立预测模型。在2006年和2007年影响浙江省的2个台风的实际预测中，主成分RBF预测能够减少台风灾害损失的误差。因此，该模型可用于实际台风灾害损失预测，有效地提高预测台风灾害损失值的可靠性，对于浙江省乃至全国防灾减灾工作有着重大的实际意义。
본연구기우1953-2007년등륙혹대절강성유중대영향적태풍역사안례수거，고필영향태풍재해손실대소적주요인소유태풍치재인자、승재체폭로성영향인자여승재체취약성영향인자，운용주성분분석법대표시승재체폭로성영향인소여승재체취약성영향인소진행수거처리，제취주성분작위RBF신경망락모형적수입，종이건립예측모형。재2006년화2007년영향절강성적2개태풍적실제예측중，주성분RBF예측능구감소태풍재해손실적오차。인차，해모형가용우실제태풍재해손실예측，유효지제고예측태풍재해손실치적가고성，대우절강성내지전국방재감재공작유착중대적실제의의。
Based on 1953-2007 landed or has a significant effect on Zhejiang typhoon case history data, Considering the main factors influencing the size of the typhoon disaster losses a typhoon to exposure to sex of hazard-affected bodies influence factor and the vulnerability of hazard-affected bodies influence factor, using the principal component analysis to said sexual exposure factors of hazard-affected bodies and the vulnerability of hazard-affected bodies influence factors in data processing, to extract the principal component as the input of RBF neural network model, This study establishes the prediction model. Impact of Zhejiang province in 2006 and 2006 two typhoons in actual forecasting, principal component analysis and RBF prediction can reduce the error of the typhoon disaster losses as a result, the model could be used to forecast actual typhoon disaster losses, and effectively improve the reliability of forecasting typhoon loss value, disaster prevention and mitigation work for Zhejiang province and even the whole country has significant practical significance.