构造了CAD系统模糊设计的一种具体解决方案:其环境为收集到的现场数据;学习环节采用基于遗传算法的模糊优化算法;知识库由设计准则构成;执行部件为设计单元.建立了回归方程的模糊优化学习算法,并构造了该算法的流程.然后利用该模糊设计系统获得了飞边尺寸设计准则,且应用实例对该算法的稳定性进行了校验.为评估该算法的性能,将其与最小二乘法和免疫遗传算法进行了比较,结果表明,该算法速度快,精度高,稳定性好.
A practical scheme of fuzzy design in CAD systems is developed,of which the environment is the currently collected data;the learning unit is the fuzzy optimization algorithm based on the genetic algorithms;the knowledge base is composed of design criteria;the executive part is the design unit.The fuzzy optimization learning algorithm of the regression equation is developed,and the corresponding flow chart is built.Then,the design criterion of a flash size is obtained by using this system;and the stability of the algorithm is verified through some examples.To evaluate the performances of the algorithm,we compare it with the least-squares method(LSM)and the immune-genetic algorithm(IGA);the result shows that our algorithm is faster,with higher precision and stability than the other algorithms.