论文部分内容阅读
针对现有机理建模算法普遍存在计算过程复杂的问题,研究了最小二乘支持向量回归机在电子设备电磁脉冲场耦合建模中的应用。先利用稳压电源在吉赫兹横电磁波室内对矩形脉冲场感应得到的耦合电压数据,对电磁脉冲能量耦合传递函数进行最小二乘支持向量回归机建模,并基于模拟退火算法、遗传算法2种典型算法对模型进行了参数优化,再运用建立的模型对耦合电压进行了仿真预测。通过对比表征模型回归能力的拟合度、均方误差2个重要参数证明,利用模拟退火算法优化的最小二乘支持向量回归机,建立的矩形脉冲场耦合模型与实际数据的拟合度更高,均方误差更小,而计算过程得到了简化,是一种简便有效的工程仿真建模手段。
Aiming at the problem that the existing algorithms are complicated, the least-squares support vector regression (LS-SVM) is used to model the electromagnetic pulse field coupling in electronic devices. Firstly, the coupling voltage data of rectangular pulse field induced by a regulated voltage power supply in the transverse electromagnetic wave cavity of a constant-voltage power supply was used to model the energy transfer coupling function of the electromagnetic pulse by least-squares support vector regression. Based on the simulated annealing algorithm and the genetic algorithm The typical algorithm optimizes the parameters of the model, and then uses the established model to simulate the coupling voltage. By comparing the fitting ability of regression model and the mean square error of the model, it is proved that the fitted model of the rectangular pulse field has a better fitting degree with the actual data by using the least-squares support vector regression machine optimized by simulated annealing algorithm. , The mean square error is smaller, and the calculation process is simplified, which is a simple and effective engineering simulation modeling method.