Advanced Power Quality Enhancement and Harmonic Filtering in Solar Photovoltaic Integrated Cascaded H-Bridge Inverters

Globally, grid-connected solar power systems are growing in popularity, but a significant challenge is optimizing output power while preserving excellent power quality. The depletion of fossil resources and the environmental issues surrounding traditional energy sources have led to a surge in interest in renewable energy, especially solar energy. This research suggests a hybrid control technique for enhancing the power quality of solar photovoltaic systems that are grid-connected by mitigating the effect of harmonic currents on system performance, even when linear loads are involved. The technique makes use of a 17-level cascaded H-bridge multilevel inverter that is coupled with the Puma optimizer, a bidirectional long short-term memory (BiLSTM) network, and an advanced multi-head cross-attention-based optimizer. The Puma optimizer enhances the deep learning network's learning process, while the BiLSTM network regulates the input to the cascaded H-bridge multilevel inverter to drive its output. This combination improves the efficiency of the photovoltaic generator and maximizes power extraction in the system. Simulations are performed in MATLAB to improve power quality and filter harmonics in solar photovoltaic systems for two cases: linear loads with and without fault conditions, with a constant level of radiation at 1000 W/m². The fundamental frequency of the 17-level cascaded H-bridge multilevel inverter is 50 Hz, with a magnitude of approximately 397.8. The overall harmonic distortion is kept at 3.96%. Additionally, the Puma optimizer has the lowest mean squared error (0.025).

Cite this article as: V. S. Pawar and P. J. Gaidhane, "Advanced power quality enhancement and harmonic filtering in solar photovoltaic integrated cascaded H-bridge inverters," Electrica, 25, 0210, 2025. doi: 10.5152/electrica.2025.24210.