基于TSK模型的模糊推理改进算法  

An Improved Fuzzy Reasoning Algorithm Based on TSK Model

作　　者：田一慧[1] 钱皓[1] 王涛[1] 

机构地区：[1]辽宁工业大学数理科学系,辽宁锦州121001

出　　处：《辽宁工业大学学报（自然科学版）》2009年第4期255-261,共7页Journal of Liaoning University of Technology(Natural Science Edition)

基　　金：辽宁省教育厅(重点实验室)基金项目(20060395)

摘　　要：在传统的基于TSK模型的模糊推理算法基础上,研究了一种改进的基于TSK模型的模糊推理新算法,并应用模糊神经BP算法给出三角形隶属函数下的算法的过程,最后将新算法与传统算法做了比较,得出基于TSK模型的模糊推理新算法在实际的过程中克服了传统推理算法会出现弱连续或不连续情况的优点。For TSK fuzzy reasoning model with two linguistic variables, two inputs and one output. If the inference antecedents is Triangular-type membership functions. By using back-propagation learning algorithm to learn and adjust the parameters of the function membership in the fuzzy reasoning rules, the conventional neuro-fuzzy reasoning algorithms and the improved neuro-fuzzy reasoning algorithms are proposed, respectively. Finally, some comparisons between these two algorithms are made. The main advantages of the improved neuro-fuzzy reasoning algorithms were that the case of weak-firing or non-firing was all avoied, which occurred from the traditional approach.

关 键 词：模糊推理 TSK模型 BP算法 神经网络 

分 类 号：O174.4[理学—数学]