Figure 3 | Scientific Reports

Figure 3

From: Deep learning-based inverse design of microstructured materials for optical optimization and thermal radiation control

Figure 3

(a–c) Emissivity spectrums of three test cases (Ideal heating, ideal cooling, and unity emissivity) input into the inverse spectrum. The reflectivity is computed as R = 1 – E, and transmissivity is set to 0. (d–f) ML generated refractive index (n) and extinction coefficient (k) for each of the test cases compared to the material properties of a material in the library that most closely matches it. (g–i) FDTD simulation results for both the ML generated material and the closest matching library material.

Back to article page