The manufacturing precision of slot in the plane slotted-array antenna has a great effect on the electrical performance. In order to avoid the shortcomings of assumptions and approximations made by some traditional methods during the course of studying the effect, a new method using support vector regression is proposed to obtain the prediction model of electrical performance affected by manufacturing precision, according to the data from the manufacturing of plane slotted-array antennas. Moreover, a method minimizing both model fitting capabilities and generalization as the objective function is offered to choose the proper parameters of support vector regression and facilitate the application. Simulation research shows that the proposed method can accurately predict the effect and has better generalization compared with BP neural network and least squares regression.%为了避免传统方法在分析结构因素对电性能影响时需要假设和近似处理的弊端,提出了根据平板裂缝天线生产制造中的数据,使用支持向量回归建立缝制造精度对电性能指标影响的预测方法.给出了一种以最小化支持向量回归的拟合能力和泛化能力为目标函数来选择支持向量回归机参数的方法,以方便该方法的工程应用.案例研究表明,该方法能够较准确地预测缝制造精度对电性能的影响;与传统的BP神经网络和最小二乘法对比,该方法具有更好的泛化能力,可以应用到平板裂缝天线的计算机辅助制造中.
展开▼
