东北林业大学学报
東北林業大學學報
동북임업대학학보
JOURNAL OF NORTHEAST FORESTRY UNIVERSITY
2014年
8期
75-78
,共4页
金森%张运林%朱凯月%胡悦
金森%張運林%硃凱月%鬍悅
금삼%장운림%주개월%호열
含水率%风速%压缩比%烟头%蒙古栎
含水率%風速%壓縮比%煙頭%矇古櫟
함수솔%풍속%압축비%연두%몽고력
Moisture content%Wind speed%Packing ratio%Cigarette end%Mongolian oak
以蒙古栎凋落叶片为实验材料,在室内构建不同含水率、风速及压缩比的可燃物床层,用烟头对蒙古栎落叶床层进行了900余次点燃燃烧试验,记录点燃概率,并进行统计分析。结果表明:风速对点燃概率有很大影响,当风速<2m? s-1时,不论蒙古栎落叶的含水率与压缩比如何改变,烟头都不能点燃蒙古栎落叶;当风速≥2m?s-1时,可燃物能够点燃,但风速对点燃概率无显著作用。含水率、床层压缩比对燃烧概率有极显著作用,3个因子间的交互作用对燃烧概率也有极显著作用。当可燃物含水率超过10%时,无论有风无风,烟头都难以点燃蒙古栎凋落叶片。用Logistic回归拟合得到含水率、风速、压缩比对烟头点燃蒙古栎落叶的概率模型,模型预报的准确率在90%左右,误报率不超过40%。
以矇古櫟凋落葉片為實驗材料,在室內構建不同含水率、風速及壓縮比的可燃物床層,用煙頭對矇古櫟落葉床層進行瞭900餘次點燃燃燒試驗,記錄點燃概率,併進行統計分析。結果錶明:風速對點燃概率有很大影響,噹風速<2m? s-1時,不論矇古櫟落葉的含水率與壓縮比如何改變,煙頭都不能點燃矇古櫟落葉;噹風速≥2m?s-1時,可燃物能夠點燃,但風速對點燃概率無顯著作用。含水率、床層壓縮比對燃燒概率有極顯著作用,3箇因子間的交互作用對燃燒概率也有極顯著作用。噹可燃物含水率超過10%時,無論有風無風,煙頭都難以點燃矇古櫟凋落葉片。用Logistic迴歸擬閤得到含水率、風速、壓縮比對煙頭點燃矇古櫟落葉的概率模型,模型預報的準確率在90%左右,誤報率不超過40%。
이몽고력조락협편위실험재료,재실내구건불동함수솔、풍속급압축비적가연물상층,용연두대몽고력락협상층진행료900여차점연연소시험,기록점연개솔,병진행통계분석。결과표명:풍속대점연개솔유흔대영향,당풍속<2m? s-1시,불론몽고력락협적함수솔여압축비여하개변,연두도불능점연몽고력락협;당풍속≥2m?s-1시,가연물능구점연,단풍속대점연개솔무현저작용。함수솔、상층압축비대연소개솔유겁현저작용,3개인자간적교호작용대연소개솔야유겁현저작용。당가연물함수솔초과10%시,무론유풍무풍,연두도난이점연몽고력조락협편。용Logistic회귀의합득도함수솔、풍속、압축비대연두점연몽고력락협적개솔모형,모형예보적준학솔재90%좌우,오보솔불초과40%。
Fuelbeds composed of Mongolian oak leave were constructed with varied fuel moisture contents and packing ratios .Over 900 ignition experiments were conducted by dropping cigarette ends on the fuelbeds to test burning probability of the fuels under varied wind speeds .The wind speed significantly influenced burning probability of the fuelbed .No fuelbed was igni-ted when wind speed was lower than 2 m? s-1 regardless of fuelbed packing ratio and moisture content .Fuebed was ignited when wind speed was higher than 2 m? s-1 , but the wind speed makes no significant differences on burning probability . Fuel moisture content , packing ration and their interaction with wind speed affected burning probability .No burning oc-curred when fuel moisture content was higher than 10%.A logistic model was established for predicting burning probability by using these variables with accuracy of 90%.