Day-Ahead Forecasting of Hourly Photovoltaic Power Based on Robust Multilayer Perception

Abstract

Photovoltaic (PV) modules convert renewable and sustainable solar energy into electricity. However, the uncertainty of PV power production brings challenges for the grid operation. To facilitate the management and scheduling of PV power plants, forecasting is an essential technique. In this paper, a robust multilayer perception (MLP) neural network was developed for day-ahead forecasting of hourly PV power. A generic MLP is usually trained by minimizing the mean squared loss. The mean squared error is sensitive to a few particularly large errors that can lead to a poor estimator. To tackle the problem, the pseudo-Huber loss function, which combines the best properties of squared loss and absolute loss, was adopted in this paper. The effectiveness and efficiency of the proposed method was verified by benchmarking against a generic MLP network with real PV data. Numerical experiments illustrated that the proposed method performed better than the generic MLP network in terms of root mean squared error (RMSE) and mean absolute error (MAE).

Islanding Fault Detection in Microgrids—A Survey

“…At this stage, the islanding fault can be detected by comparing the average voltage unbalance given in (10) with the reference value VU r . As suggested in Reference [57], the comparison should be done every 4.17 [ms] (i.e., one quarter of a cycle if the nominal frequency is 60 [Hz]).…”

“…However, numerous problems should be addressed before their widespread usage in the power networks. These problems include frequency stabilization (e.g., model predictive control [1], passivity-based approach [2], and Lyapunov-based approach [3]), voltage stabilization (e.g., PI controller [4], fuzzy logic [5], droop control [6], and consensus-based approach [7]), robustness against uncertainties (e.g., optimization-based approach [8], load shedding-based approach [9], neural networks [10], and three phase improved magnitude phase locked loop [11]), and tolerance toward faults (e.g., reconfiguration-based approach [12], model predictive control [13], optimization-based approach [14], model-based techniques [15], and a consensus-based approach [16]).…”

Whistleblowing on photovoltaic operations in Nigeria: panacea for sustainable development

“…This has prompted most local business organizations to subscribe to the hybrid system, which comprises PV coupled to battery and generator, to sustain operations and profitability (Oparaku 2003;Adesanya and Pearce 2019). Noteworthy is that variability of PV power poses a challenge to its grid operation and utilization (Ming et al 2017;Huang et al 2018;Ebhota and Jen 2019).…”

Robust 24 Hours ahead Forecast in a Microgrid: A Real Case Study

“…In [11], a hybrid forecasting algorithm was proposed, based on ANN and fuzzy logic pre-processing, in order to increase forecast accuracy. In particular, the robustness of the ANN approach for day ahead PV forecasting was also assessed in [12]. In [13], a hybrid ANN model for the PV power forecasting exploiting clear-sky models and ANN ensembles, based on day ahead weather forecast, was validated on a real PV plant as a robust and accurate procedure.…”