RESUMEN
In this work, we demonstrate the training, conversion, and implementation flow of an FPGA-based bin-ratio ensemble spiking neural network applied for radioisotope identification. The combination of techniques including learned step quantisation (LSQ) and pruning facilitated the implementation by compressing the network's parameters down to 30% yet retaining the accuracy of 97.04% with an accuracy loss of less than 1%. Meanwhile, the proposed ensemble network of 20 3-layer spiking neural networks (SNNs), which incorporates 1160 spiking neurons, only needs 334 µs for a single inference with the given clock frequency of 100 MHz. Under such optimisation, this FPGA implementation in an Artix-7 board consumes 157 µJ per inference by estimation.
Asunto(s)
Redes Neurales de la Computación , Neuronas , Neuronas/fisiología , Potenciales de Acción/fisiología , Radioisótopos , Algoritmos , HumanosRESUMEN
The applicability and performance of FeS in ozonation process to remove p-aminobenzenesulfonamide (SN) from water was assessed, and the working mechanism of FeS was comprehensively explored by both experimental means and density functional theory (DFT) simulation. FeS combined ozonation achieved 74% of SN removal in 60 min under the optimal condition, which was 37% higher than by ozonation alone, and 12% higher than FeO combined ozonation. Highly active species of â¢OH, â¢SO4-, 1O2 and â¢O2- were detected in the FeS combined ozonation system, the evolution pathway of the involved species was expounded with the aid of DFT calculation. The results revealed that â¢O2-, H2O2 and SO42- were originally formed via interface reactions on FeS surface, then gradually transformed into â¢OH, 1O2 and â¢SO4- through subsequent chain reactions. Moreover, FeS had a lower energy barrier of 0.16 eV than FeO with a value of 0.83 eV for the transformation of ozone to active atomic oxygen. The presented study provided a significant insight into the role of Fe-based materials in ozonation, and was of great importance to guide the route for ozonation process improvement.