Figure 4 | Scientific Reports

Figure 4

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

Figure 4

(a–f) Narrowband simulation results using the inverse neural network. The input spectrum has an emissivity of 0.05 throughout the λ = 0.3 to 16 um, except for two points that define the “peak” location which have an emissivity of 1. The reflectivity is defined by E = 1 – R and the transmission is set to 0. The results are compared to the desired input as well as the surrogate predictions for the same material.

Back to article page