华北水利水电大学学报(自然科学版)
華北水利水電大學學報(自然科學版)
화북수이수전대학학보(자연과학판)
Journal of North China University of Water Resources and Electric Power
2015年
5期
10-13
,共4页
唐言明%卜松%董洪茂%范江洋
唐言明%蔔鬆%董洪茂%範江洋
당언명%복송%동홍무%범강양
灾情评估%信息熵%遗传算法%指数普适公式
災情評估%信息熵%遺傳算法%指數普適公式
재정평고%신식적%유전산법%지수보괄공식
disaster assessment%information entropy%genetic algorithms%index universal formula
作为区域灾害风险管理的重要组成部分,灾情评估可为区域灾害预防与发生以及更好地进行区域经济可持续发展规划提供科学依据.自然灾害灾情评估的难点是估算灾情发生的不确定性、损失的概率分布以及处理评价过程中指标的随机性和评价等级的模糊性.为此,提出了基于信息熵的改进灾情评估模型,用最大熵原理处理灾害发生的不确定性和损失的概率分布,用指数普适公式处理评价指标的变化差异性,并利用加速遗传算法优化相关参数,进而建立了基于信息熵的AGA优化改进灾情评估普适模型.应用结果表明:该模型有效地估算了灾害发生的概率,解决了评估灾害发生的不确定性问题,提高了评价的可靠性,且方法智能化程度高,可在多种灾情评价中进行推广应用.
作為區域災害風險管理的重要組成部分,災情評估可為區域災害預防與髮生以及更好地進行區域經濟可持續髮展規劃提供科學依據.自然災害災情評估的難點是估算災情髮生的不確定性、損失的概率分佈以及處理評價過程中指標的隨機性和評價等級的模糊性.為此,提齣瞭基于信息熵的改進災情評估模型,用最大熵原理處理災害髮生的不確定性和損失的概率分佈,用指數普適公式處理評價指標的變化差異性,併利用加速遺傳算法優化相關參數,進而建立瞭基于信息熵的AGA優化改進災情評估普適模型.應用結果錶明:該模型有效地估算瞭災害髮生的概率,解決瞭評估災害髮生的不確定性問題,提高瞭評價的可靠性,且方法智能化程度高,可在多種災情評價中進行推廣應用.
작위구역재해풍험관리적중요조성부분,재정평고가위구역재해예방여발생이급경호지진행구역경제가지속발전규화제공과학의거.자연재해재정평고적난점시고산재정발생적불학정성、손실적개솔분포이급처리평개과정중지표적수궤성화평개등급적모호성.위차,제출료기우신식적적개진재정평고모형,용최대적원리처리재해발생적불학정성화손실적개솔분포,용지수보괄공식처리평개지표적변화차이성,병이용가속유전산법우화상관삼수,진이건립료기우신식적적AGA우화개진재정평고보괄모형.응용결과표명:해모형유효지고산료재해발생적개솔,해결료평고재해발생적불학정성문제,제고료평개적가고성,차방법지능화정도고,가재다충재정평개중진행추엄응용.
As an important part of regional disaster risk management, damage assessment can provide a scientific basis for the regional disaster prevention and occurrence and better planning for sustainable development of regional economy. The difficulties of natural dis-aster assessment are the estimates of the disaster uncertainty and the probability distributions of the loss, and the randomness of indica-tors and the ambiguity of evaluation level in the evaluation process. Therefore, an improved disaster assessment model based on infor-mation entropy is proposed. The maximum entropy principle is applied to the disaster uncertainty and the probability distributions of the loss, the index universal formula is applied to the variation differences of evaluation indexes, and the accelerating genetic algorithm is used to optimize the relevant parameters. Furthermore, an AGA optimized improved universal model of flood disaster assessment based on the information entropy is built. The application results show this model effectively estimates the probability of disaster, solves the uncertainty of disaster occurrence, and improves the reliability of evaluation. Thus, the method is smart and effective, and can be per-formed in a variety of disaster evaluation.