基于遗传神经网络的地下商场火灾风险评价方法-安全技术及工程专业论文.docxVIP

Subject：The Subject：The Method of Fire Risk Evaluation of Underground Emporium Based On Genetie Neural Networks Specialty：Safety Technology and Engineering Name：Yue Hailing (Signature) Instructor：Li Shugang (Signature) Abstract Underground emporiums construction structure special，the fire risk is big，the difficulty of scapes and saves goal to be big，so it needs regularly to carry on the present situation evaluation．But now our country lakes of evaluation methods underground emporiums this kind of special place．Based on the above reasons，this paper chooses underground emporiums fire risk present situation evaluation method to research，has the vital significance regarding the large—scale public building fire risk evaluation as well as the performance design． Firstly,this paper according to the underground emporiums existing stastus and each kind of building standard as wel／as the fire reason statistics，by the study of fire’s happeness and development，we analyzed comprehensively the hazard factor of underground emporiums． On the basis of adhering to principle of safety assessment，established the evalutative index system suitable for the underground emporiums by causal analytical method．Then three kinds of dangerous sources law the paper sets up hazard safety assessment index that is full represent major influencing factors ofthree—class ofhazard． Based on the complementation of advantages of GA and NN，we probeinto the necessity and feasibility of their conjunction and associate them together,a new mixing GANN model which possesses the advantages both GA and NN isestablished．Matlab，oue of the programming assistant which is simple，directly,efficient and open is selected to fulfill this model． On the base of collected data and expels’advice，this paper uses the grey clustering is suitable insufficient samples to create the training genetic nerve network samples．Then designed particular each parameters of genetic neural network，the network runconsult