论文部分内容阅读
本文对一类直接可观测的双线性系统用沃尔什函数方法进行辨识的问题,进行了一些探讨,把原来的沃尔什函数辨识方法改进为一种迭代算法,从而较好地提高了系统辨识的精确度。通过数字仿真,讨论了函数用沃尔什级数近似的项数以及采用不同的输入信号对辨识精确度的影响,验证了这种辨识方法在有噪声存在时的有效程度。为了验证这种方法用于实际系统的可能性,对一个实际的热交换器进行了辨识,结果表明,对于那些近似于双线性的实际过程,沃尔什函数迭代算法是一种有效的辨识方法、
In this paper, the problem of recognizing a class of directly observable bilinear systems using the Walsh function method is discussed, and the original Walsh function identification method is improved as an iterative algorithm, thus improving the System identification accuracy. Through numerical simulation, the number of items approximated by the Walsh series of the function and the influence of different input signals on the recognition accuracy are discussed. The effectiveness of this method in the presence of noise is verified. To verify the feasibility of this method for practical systems, an actual heat exchanger was identified and the results show that the Walsh function iterative algorithm is an effective method for those bilinear approximations method,