A REVIEW: POWER QUALITY MEASURMENT IN SOLAR PHOTOVOLTIC SYSTEM USING ANN CONTROLLER
SAPNA XALXO
Mtech Scholar
Dept of Electrical, RSR Rungta college of engineering and technology
MANISH KUMAR CHANDRAKAR
Assistant Professor
Dept of Electrical, RSR Rungta college of engineering and technology
Abstract:
Power electronics and D.E.R. have contributed to rising power quality problems on the utility grid. The power quality must be kept constant under both steady and fault circumstances, regardless of the load. Distribution grid power quality may be enhanced with FACTS's Unified Power Quality Conditioners. The reactive and actual power imbalances are corrected by the Unified Power Quality Conditioner at the same time. Combining reactive power control and unit vector template control is the proposed control algorithm for the Unified Power Quality Conditioner. By combining the Unified Power Quality Conditioner with D.E.R.s like solar panels, converter power ratings may be lowered while still meeting demand. When applied to power electrical devices, reinforcement learning algorithms, in particular the Neural-Network method, can significantly boost efficiency. Superior power generation is achieved by using an Artificial Neural Network controller in a solar-integrated Unified Power Quality Conditioner system. The controller can adjust to new conditions on its own. The system is put through its paces with balanced and unbalanced loads in MATLAB-SIMULINK. The per-unit method simplifies analysis. Non-linear load distortion can be eliminated by solar integration at the DC link and hybrid management of the series and shunt converters. Short voltage dips and spikes are needed for IEEE 1159. In just 50 ms, the Computer and Business Equipment Manufacturers Association curve glides over the brief dip or spike that can last anywhere from half a second to three. The suggested control mechanism naturally reduced harmonics in the load-side current. Both power factor and distribution dependability are protected in this way. An artificial neural network controller in a solar integrated Unified Power Quality Conditioner was shown to be superior to a traditional Proportional-Integral controller in reducing harmonics of varying orders by 12.72 percentage points. Total Harmonic Distortion was lowered to below IEEE 519 limits thanks to a controller based on artificial neural networks. The goal of achieving several harmonic orders was met.
Keywords—solar radiation prediction; solar energy modeling;artificial intelligence; artificial neural networks(ANN)