【摘要】 该文提出一种非线性系统建模方法。该方法包含 3个部分 :独立输入信号的标准正交分解 ,从输入信号的权函数中导出标准Wiener泛函级数 ,由泛函级数线性组合成系统模型 ,利用最小二乘优化组合系数。最后列举实例 ,根据标定的数据用该方法对某传感器进行建模 ,结果验证了该建模方法对此类非线性结构的预测和辨识性能。更多还原

【Abstract】 A nonlinear system modeling scheme was proposed. This scheme consists of three sections: a signal-independent standard orthonormal expansion,standard Wiener functional series derived from an estimate of vector probability density function, and a linear combiner of functional series. The prediction and identification performance of these nonlinear structures were examined by an example, which models a sensor according to the results of calibration.更多还原