FFM功能求解及在传感器控制系统中的应用  

FFM function-solving and its application in control system for sensor

作　　者：华丹阳[1,2] 刘晓平[1] 唐益明[1] 

机构地区：[1]合肥工业大学计算机与信息学院,合肥230009 [2]安徽农业大学信息与计算机学院,合肥230036

出　　处：《电子测量与仪器学报》2016年第6期975-981,共7页Journal of Electronic Measurement and Instrumentation

基　　金：国家自然科学基金(61203077;61370167);中国博士后科学基金(2014T70585;2012M521218);安徽省自然科学基金(1408085MKL15;1508085QF129)项目资助

摘　　要：传统产品概念设计的功能树中功能节点取值非0即1,且不同方案完成相同功能时无法自动判优,针对这些问题提出了新的功能求解思路。首先,引入模糊表示理论,定量地刻画出功能节点模糊值,构造了模糊功能树。其次,提出了模糊功能矩阵,构建了展开定理和优选定理,通过交替展开与优选,建立了FFM功能求解算法。最后,将该算法应用于传感器概念设计中,通过与组合求解算法、扩展功能矩阵的约简算法进行对比,验证了所提算法的有效性和优越性。通过研究发现:所提算法能够更加细致地刻画功能的粒度,可以对实现相同功能的类似方案进行自动判优,在保证了创新能力不损失的前提下,提高了功能求解的效率和质量,从而推动了概念设计领域的发展。In traditional product concept design,the value of function tree node is 0 or 1,and there lacks automatic arbitration for different schemes which complete the same function. Aiming at these problems,a novel functionsolving idea is proposed. First of all,fuzzy expression theory is introduced,and the fuzzy values of function nodes are provided. And then,the fuzzy function tree is constructed. Furthermore,Fuzzy Function Matrix（ FFM） is put forward,while the expanding theorem and arbitration theorem are proposed,and thus FFM function-solving algorithm is established with alternate expanding and arbitration. Finally,by the application in sensor conceptual design,the proposed algorithm is found to be effective and better than the combinational solving algorithm and the reducing method via extended function matrix. It is finded by research,the proposed algorithm can be more meticulous to characterize the granularity of function,and carry on automatic arbitration for different schemes with the same function,and effectively enhance the function-solving ＇s efficiency and quality without decreasing the innovation capability,which improves the development of conceptual design.

关 键 词：概念设计 模糊表示 功能矩阵 传感器 

分 类 号：TP391.72[自动化与计算机技术—计算机应用技术]