Enhancing Power Quality through Harmonic Reduction: A Triple-Memristor Hopfield Neural Network Approach
K S B Varaprasad, Y. Sukhi, Shanmugapriya Subramaniyan, Ramya Dhandapani, M. Rajendiran, Seeniappan Kaliappan · 2024
Harmonic distortion in power systems is a significant concern that impacts the efficiency, reliability, and safety of electrical networks. Implementing effective harmonic distortion mitigation solutions, such as filters or advanced control systems, can be expensive and may necessitate substantial changes to existing infrastructure. This paper proposes a Triple-Memristor Hopfield Neural Network (TMHNN) as a novel method for mitigating harmonic distortion in power systems. The main objective of the proposed approach is to improve the quality of power by mitigating the harmonics. The TMHNN is used to predict the actual harmonic current injected by specific harmonic sources within a power system. The results demonstrate the significant advantages of the proposed TMHNN method over existing techniques such as Artificial Neural Network (ANN), Adaptive Neuro-Fuzzy Inference System (ANFIS), and Recurrent Neural Network (RNN). TMHNN achieves the highest efficiency at 96.3%, lowest error rate at 4.16% and minimizing total harmonic distortion (THD), with the lowest value of 3.74%. These results highlight TMHNN's superior performance in terms of efficiency, harmonic distortion reduction, and accuracy.