【摘要】 为更好地消除家庭厨房中的火灾安全隐患，便于家庭成员进出厨房，推动对智能化建筑的发展，设计研究了一种智能安全防护厨房门，能够对厨房火灾进行有效的预防。针对传统厨房中火灾报警器决策因子单一从而导致厨房火灾预警系统识别精度低、可靠性差等问题，利用厨房门上多个传感器检测的特征参数，基于权值修正的D-S证据理论提出厨房火灾特征融合决策算法。该算法通过结合采集的温度、CO和烟雾浓度构建出各参数基本概率分配函数，根据三个火灾特征参数之间的关联性进行权值修正，后对各参数基本概率分配函数进行融合，得出决策。实验结果表明，当在智能安防厨房门上的多个传感器监测结果决策不一致时，通过该算法对三个火灾阶段传感检测数据进行加权融合，得到更为精准的火情决策，提高了系统的识别精度和可靠性。更多还原

【Abstract】 In order to better eliminate the hidden danger of fire safety in the family kitchen,facilitate family members to enter and exit the kitchen,and promote the development of intelligent buildings,an intelligent safety protection kitchen door is designed and studied,which can effectively prevent kitchen fires. Aiming at the problems of low recognition accuracy and poor reliability of kitchen fire early warning system caused by single decision factor of fire alarm in traditional kitchen,a kitchen fire feature fusion decision algorithm based on D-S evidence theory with weight correction is proposed using the feature parameters detected by multiple sensors on the kitchen door. The algorithm constructs the basic probability distribution function of each parameter by combining the collected temperature,CO and smoke concentration. The weight value is modified according to the correlation between the three fire characteristic parameters,and then the basic probability distribution function of each parameter is fused to obtain the decision. The experimental results show that when the decisions of the monitoring results of multiple sensors on the intelligent security kitchen door are inconsistent,the algorithm is used to fuse the sensor detection data of the three fire stages with weight,and more accurate fire decision-making is obtained,which improves the recognition accuracy and reliability of the system.更多还原