Modeling and simulation platform of optical teaching experiments SeeLight was proposed. Its modeling range covers a variety of simple or complex optical systems including geometrical optics, wave optics, adaptive optics, information optics, polarization analysis, and so on. It has seven categories of models and more than 50 simulation application examples, so that many optical teaching experiments can be done on this platform. The platform is based on the component-based software architecture. And the operating efficiency is greatly improved based on high-performance computing framework. And it’s not only a stand-alone version, but also web version, which provides convenience for the user to build optical simulation systems. So that the complex optical experiment can be easily implemented on the platform, and the parameters of the experimental system can be adjusted flexibly, which greatly reduce the experimental cost and improve the efficiency of the experimental analysis. And in the process of simulation, not only the results of all optical components which have different spatial locations can be observed, but also users can observe the results over time. And the high-performance computing framework greatly improves the efficiency of the platform. Multiple optical simulation systems of comparative experiments can be simultaneously simulated. The functional structure, typical features and key simulation difficulties of this platform were discussed in detail. Finally, several simulation examples were given.
Propagation simulation method and choosing mesh grid are both very important to get the correct propagation results in wave optics simulation. A new angular spectrum propagation method with alterable mesh grid based on the traditional angular spectrum method and the direct FFT method is introduced. With this method, the sampling space after propagation is not limited to propagation methods no more, but freely alterable. However, choosing mesh grid on target board influences the validity of simulation results directly. So an adaptive mesh choosing method based on wave characteristics is proposed with the introduced propagation method. We can calculate appropriate mesh grids on target board to get satisfying results. And for complex initial wave field or propagation through inhomogeneous media, we can also calculate and set the mesh grid rationally according to above method. Finally, though comparing with theoretical results, it’s shown that the simulation result with the proposed method coinciding with theory. And by comparing with the traditional angular spectrum method and the direct FFT method, it’s known that the proposed method is able to adapt to a wider range of Fresnel number conditions. That is to say, the method can simulate propagation results efficiently and correctly with propagation distance of almost zero to infinity. So it can provide better support for more wave propagation applications such as atmospheric optics, laser propagation and so on.
In the spectral imaging process of interference imaging spectrometer, the impact caused by
motion errors of satellite platform is not only the degradation of spatial resolution but also the aliasing and
distortion of spectral information. The degradation function is modeled though analyzing the degradation
mechanism of spectral imaging. And a correction algorithm based on interference sequences is proposed to
solve this problem. In this algorithm, the interference sequences is regarded as an integer and the degradation
model is thought to be three-dimensional, including two spatial dimensions and one spectral dimension. And
the degraded interference sequences are the convolution of the original sequences and the degradation
function. According to the vibration data measured by POS on the satellite, the degradation function can be
obtained. Then the corrected interference sequences can be got by de-convolution of the degraded sequences
and the degradation function though Hopfield neural network based on continuous variation of states. And by
reconstructing of the corrected sequences, the corrected images and spectrums can be got finally. The
simulated results show that this algorithm can correct the degradation of images and the distortion of
spectrums caused by motion errors of satellite's attitudes effectively.