The invention discloses a BP neural network-based spliced
telescope translation aberration detection technology, belongs to the technical field of
active optics, and aims to solve the problems of highhardware cost,
complex calculation and low efficiency of the conventional spliced
telescope translation aberration detection technology. The method comprises the following steps: 1, establishing an accurate theoretical relational expression of a sidelobe module value and sub-mirror translation aberration; 2, constructing a BP neural network, and training an
artificial neural network model of thedistorted far-field
light intensity image and the sub-mirror translation aberration by using the established
data set; 3, aiming at the specific spliced
telescope system in the step 2, acquiring lightintensity image information of a
point source observation target on a focal plane under the same
broadband spectrum light source, and inputting the
light intensity image information into a main control computer; and 4, enabling the main control computer to firstly perform
Fourier transform on the focal plane image, extract a sidelobe module value corresponding to each sub-mirror of a
Fourier transform function as the input of a
network model, and directly output the translation aberration of each sub-mirror by using the BP neural
network model constructed in the step 2.