Harmonic detection technology based on ant colony optimization BP neural network

Abstract

At present, ip-iq algorithm based on instantaneous reactive power theory is the most widely used in the field of harmonic detection.However, the ip-iq algorithm based on instantaneous reactive power has coordinate transformation and inverse transformation, which leads to cumbersome calculation.the existence of low-pass filter leads to delay and poor dynamic performance of the algorithm.The harmonic detection based on BP neural network has the advantages of simple calculation, fast detection speed, strong adaptive ability and good dynamic performance.However, the initial weights and thresholds are random, so it is easy for the network to fall into local optimal solution during training, resulting in low detection accuracy. This paper presents a harmonic detection method based on ant colony algorithm optimized BP neural network. ACO-BP, BP and ip-iq algorithms are compared by Simulink simulation. The results show that ACO-BP harmonic detection algorithm is feasible and superior

From Patents to Progress: Genetic Algorithms in Harmonic Distortion Monitoring Technology

“…In the first group, we identified applications capable of monitoring and predicting future demand for quality electricity distribution (CN201610894806A) [79], as well as planning and improving supply in distribution networks (CN202110735473A, CN202110859394A, CN201910635250A) [70,82,85]. On the other hand, in the second group, most of the patents seek to solve the problem by optimizing network distribution based on improving detection (CN201710783960A) [81], reducing voltage losses (CN202211270838A) [80] in microgrids (CN202110688202A) [84], and suppressing harmonics through specific devices and methods (CN202210645076A) [83], thus improving power quality and the reliability of power supply.…”