Extended Stochastic Gradient Identification Method for Hammerstein Model Based on Approximate Least Absolute Deviation

Table 2

Identification results with different .

Parameters	a1	a2	b1	b2	c2	c3	d1	(%)
	−1.5711	0.7832	0.8461	0.7040	0.4769	0.2501	−0.7541	10.2411
	−1.5177	0.7147	0.8189	0.7741	0.4764	0.2580	−0.7038	11.4612
	−1.4660	0.6617	0.8396	0.7817	0.4818	0.2515	−0.6394	15.1781
True value	−1.600	0.800	0.850	0.650	0.500	0.250	−0.640	0