Modeling bidirectional reflection distribution function of microscale random rough surfaces

The radiative properties of three different materials surfaces with one-dimensional microscale random roughness were obtained with the finite difference time domain method(FDTD) and near-to-far-field transformation.The surface height conforms to the Gaussian probability density function distribution.Various computational modeling issues that affect the accuracy of the predicted properties were discussed.The results show that,for perfect electric conductor(PEC) surfaces,as the surface roughness increases,the magnitude of the spike reduces and eventually the spike disappears,and also as the ratio of root mean square roughness to the surface correlation distance increases,the retroreflection becomes evident.The predicted values of FDTD solutions are in good agreement with the ray tracing and integral equation solutions.The overall trend of bidirectional reflection distribution function(BRDF) of PEC surfaces and silicon surfaces is the same,but the silicon's is much less than the former's.The BRDF difference from two polarization modes for the gold surfaces is little for smaller wavelength,but it is much larger for the longer wavelength and the FDTD simulation results agree well with the measured data.In terms of PEC surfaces,as the incident angle increases,the reflectivity becomes more specular.

A Stochastic Gradient Descent Method for Computational Design of Random Rough Surfaces in Solar Cells

In this work,we develop a stochastic gradient descent method for the computational optimal design of random rough surfaces in thin-film solar cells.We formulate the design problems as random PDE-constrained optimization problems and seek the optimal statistical parameters for the random surfaces.The optimizations at fixed frequency as well as at multiple frequencies and multiple incident angles are investigated.To evaluate the gradient of the objective function,we derive the shape derivatives for the interfaces and apply the adjoint state method to perform the computation.The stochastic gradient descent method evaluates the gradient of the objective function only at a few samples for each iteration,which reduces the computational cost significantly.Various numerical experiments are conducted to illustrate the efficiency of the method and significant increases of the absorptance for the optimal random structures.We also examine the convergence of the stochastic gradient descent algorithm theoretically and prove that the numerical method is convergent under certain assumptions for the random interfaces.