【作者】 赵恒； 方素平；

【Author】 ZHAO Heng,FANG Su-ping(The State Key Lab for Manufacturing Systems Engineering,Xi’an Jiaotong University,Xi’an 710049,China)

【机构】 西安交通大学机械制造系统工程国家重点实验室；

【摘要】 分析以往导光板网点设计模型方法的不足,提出了一种采用自适应神经模糊推理系统(ANFIS)建立网点模型的新方法。将网点半径和距光源距离作为模型输入,输出面光强为模型输出。模型参数通过混合算法学习训练,模型隶属度函数以及模糊规则通过对大量已知数据的学习得到。利用ANFIS自学习和逻辑推理能力,实现网点参数的优化,并将优化模型用于网点仿真计算。与常规策略相比,ANFIS模型具有自动调整模型规则和参数,精度高和自适应能力强的特点。更多还原

【Abstract】 This paper proposes an ANFIS strategy for generation of the dot patterns in the light guide aiming to the shortcoming of the dot pattern design methods in the past.During the process of generating dot patterns,two inputs are the dot radius and the distance form dots to a light source,and one output is the luminance of the light guide panel.The parameters are trained by a hybrid algorithm,while the membership functions and rules are obtained through data learning.These parameters are optimized by using ANFIS’s self-training and logical inference.Experiment results reveal that the proposed strategy can achieve an even luminance condition compared to the common strategy.In addition,it can adjust the rules and parameters,and has the high accuracy.更多还原