Optimization of Power Communication Transmission Network Based on Nonlinear Effects

In response to the increasingly prominent nonlinear effects in power communication transmission networks, which severely limit transmission distance and capacity, the author proposes a multi-level and scalable nonlinear optimization method system. This system covers path selection, optical amplifier configuration, nonlinear equalization, and intelligent scheduling. Based on the collaborative simulation platform of MATLAB and OptiSystem, typical fiber optic models such as G.652 were selected to construct multiple network topologies and channel conditions. The system tested the improvement effect of the proposed method in terms of bit error rate (BER), throughput, and resource utilization. The experiment shows that the author’s method reduces the BER to 2.3 Å~ 10−5 at a distance of 600 km and increases the transmission distance to 990 km (an increase of 219.4% compared to no compensation); in a multi-node network, the average throughput is increased by about 40%, and the resource utilization rate reaches 82.3%. Among the intelligent algorithms compared, reinforcement learning achieved the best overall results in terms of convergence speed, BER suppression, and dynamic adaptability. Research has shown that this method has good engineering practicality and promotional value. Cite this article as: Y. Liu, K. Dong, Y. Quan and H. Liu, “Optimization of power communication transmission network based on nonlinear effects,” Electrica, 2026, 26,0231, doi: 10.5152/electrica.2026.25231.