Design and optimization of photovoltaic system with a week ahead power forecast using autoregressive artificial neural networks. (2022)

• Record Type: Journal Article
• Authors: Mughal, Shafqat Nabi; Sood, Yog Raj; Jarial, R.K.
• Is Part Of: Materials today. Volume 52:Part 3(2022)
• Page Start: 834
• Page End: 841
• Publication Date: 2022
• Subjects: Photovoltaics -- Forecasting -- Optimization -- Design; Materials science -- Congresses -- Periodicals; 620.1
• DOI: 10.1016/j.matpr.2021.10.223
• Languages: English
• ISSNs: 2214-7853

Abstract

Abstract: In this paper, an optimized design procedure for designing Photovoltaic system to generate power outputs is proposed. The power outputs generated using different design procedures depicted the variation in power produced which will work as a yardstick in deciding the choice of design procedure by the electric utilities. In addition, an Autoregressive Neural Network model to provide a week ahead forecast of PV Power output is also proposed. The proposed model is trained using Levenberg-Marquardt optimizer and the performance is validated using Mean absolute percentage error (MAPE). The forecast power output will be used by the energy operators in the correct dispatch of energy. The MAPE of the proposed model was found around 5.44. Further, the proposed model was compared with various time-series models developed using Waikato Environment for Knowledge Analysis (WEKA) which depicted the best performance by our method.