Optical-electromagnetic compatible devices are urgently required in intelligent building monitors and cross-band protection. Meanwhile, the insufficient systematicness and semi-empirical attempts significantly limit the prosperity of cross-band materials, causing enormous challenges for deviceization and material database construction. Herein, the systematical component-deviceization-machine learning prediction-array construction strategy is attempted to solve the bottleneck issues. A luminance-triggered camouflage-monitoring-protection triune integrated modular unit (IMU) is hierarchically encapsulated to simultaneously achieve efficient anti-electromagnetic interference (EMI), light-absorbing, quick gradient-colorization response. Moreover, an illumination intensity dataset and a surrogate model based on fully connected neural network fitting (FCNN-fitting) are constructed, which accurately predicts the light-absorbing property of IMUs and can be instructional for material selection. The IMUs are specifically assembled into a 4*4 array, aiming at multi-scenario application of programmable display, camouflage pattern, surface conformality, and rapid replaceability. This work paves the path and provides a promising strategy for optical-electromagnetic compatibility and material genetics-deviceization-array systematization.
Read full abstract