Chapter 8 - Overview of Solar-Forecasting Methods and a Metric for Accuracy Evaluation
Abstract
This chapter provides an introduction to and overview of the subsequent chapters, which discuss specific solar-forecasting technologies and time horizons. Solar-forecasting methods are classified by technique, time horizon, and application. Advantages and disadvantages of deterministic and stochastic forecasting approaches are laid out and discussed in the context of solar forecasting based on numerical weather prediction, satellite data, and ground measurements. Metrics to evaluate solar-forecasting techniques are then presented and a time horizon–invariant metric is introduced that allows comparing forecast errors across time horizons, geographical regions, and time steps. Finally, the metric is demonstrated with hour-ahead forecasts based on stochastic-learning and satellite cloud-motion vector techniques.
References (0)
Cited by (140)
Intraday probabilistic forecasts of surface solar radiation with cloud scale-dependent autoregressive advection
2023, Applied EnergySolar energy supply is usually highly volatile, limiting its integration into the power grid. Accurate probabilistic intraday forecasts of solar resources are essential to increase the share of photovoltaic (PV) energy in the grid and enable cost-efficient balancing of power demand and supply. Solar PV production mainly depends on downwelling surface solar radiation (SSR). By estimating SSR from geostationary satellites, we can cover large areas with high spatial and temporal resolutions, allowing us to track cloud motion. State-of-the-art probabilistic forecasts of solar resources from satellite imagery account only for the advective motion of clouds. They do not consider the evolution of clouds over time, their growth, and dissipation, even though these are major sources of forecast uncertainty. To address the uncertainty of cloudiness evolution, we present SolarSTEPS, the first optical-flow probabilistic model able to simulate the temporal variability of cloudiness. We demonstrate that forecasting the autocorrelated scale-dependent evolution of cloudiness outperforms state-of-the-art probabilistic advection-based forecasts by 9% in continuous ranked probability score (CRPS). This corresponds to an extension of the forecast lead time by about 45 min at constant CRPS. Our work is motivated by the scale-dependent predictability of cloud growth and decay. We demonstrate that cloudiness is more variable in time at smaller spatial scales than at larger ones. Specifically, we show that the temporal autocorrelation of cloudiness is related to its spatial scale by an inverse power law. We also demonstrate that decomposing cloudiness into multiple spatial scales in the forecasts further improves the forecast skill, reducing the CRPS by 10% and the RMSE by 9%.
Predictability and forecast skill of solar irradiance over the contiguous United States
2023, Renewable and Sustainable Energy ReviewsCurrent solar forecast verification processes place much attention on performance comparison of a group of competing methods. However, forecast verification ought to further answer how the best method within the group performs relative to the best-possible performance which one can attain under that forecasting situation, which makes the quantification of predictability and forecast skill immediately relevant. Unfortunately, the literature on the quantification of relative performance of solar irradiance has hitherto been lacking, and very few studies have focused on the spatial distributions of predictability and forecast skill of solar irradiance. The predictability and forecast skill of an atmospheric process depend on two concepts: (1) the growth of initial error in unresolved scale of motion, and (2) the forecast performance of the standard of reference. Based upon this formalism, predictability and forecast skill of solar irradiance in the United States are quantified and mapped. Through this study, a couple of common misconceptions in regard to irradiance predictability are refuted, and the original formulation of skill score revived.
Deep learning for intra-hour solar forecasting with fusion of features extracted from infrared sky images
2023, Information FusionThe increasing penetration of solar energy leaves power grids vulnerable to fluctuations in the solar radiation that reaches the surface of the Earth due to the projection of cloud shadows. Therefore, an intra-hour solar forecasting algorithm is necessary to reduce power instabilities caused by the impact of moving clouds on energy generation. The most accurate intra-hour solar forecasting methods apply convolutional neural networks to a series of visible light sky images. Instead, this investigation uses data acquired by a novel infrared sky imager on a solar tracker, which is capable of maintaining the Sun in the center of the images throughout the day and, at the same time, reducing the scattering effect produced by the Sun’s direct radiation. In addition, infrared sky images allow the derivation and extraction of physical cloud features. The cloud dynamics are analyzed in sequences of images to compute the probability of the Sun intercepting air parcels in the sky images (i.e., voxels). The method introduced in this investigation fuses sky condition information from multiple sensors (i.e., pyranometer, sky imager, solar tracker, weather station) and feature sources using a multi-task deep learning architecture based on recurrent neural networks. The proposed deterministic and Bayesian architectures reduce computation time by avoiding convolutional filters. The proposed intra-hour solar forecasting algorithm reached a forecast skill of 18.6% with a forecasting horizon of 8 min. Consequently, the proposed intra-hour solar forecasting method can potentially reduce the operational costs of power grids with high participation of solar energy.
Power quality daily predictions in smart off-grids using differential, deep and statistics machine learning models processing NWP-data
2023, Energy Strategy ReviewsMicrogrid autonomous networks need an effective plan and control of power supply, energy storage, and retransmission. Prediction and monitoring of power quality (PQ) along with efficient utilization of Renewable Energy (RE) is unavoidable to optimize the system performance without abnormalities. Alterations and irregularities in PQ must remain within the prescribed norm ranges and characteristics to allow fault-tolerant operation of the detached system in various modes of attached equipment. The PQ data for all possible combinations of grid-attached household appliances and different inside/outside conditions cannot be measured completely or described exactly by physical equations. PQ predictions on a daily basis using Artificial Intelligence (AI) models are needed because atmospheric fluctuations and anomalies in local weather with uncertainties in system states primarily influence the induced power and operation of real off-grids. A novel soft-computing method using Differential Learning, which allows modelling of complex dynamics of weather-dependent systems, is presented and compared with the recent standard deep and probabilistic machine learning. The AI models were evolved using weather data and the binary status of attached equipment in the test predetermined daily training periods. Daily statistical models process 24-h forecast data and definition load series of trained input variables to calculate the target PQ parameters at the same times. Optimal utilization, efficiency, and failure-free operation of smart grids can be planned according to the suggested operable power consumption scenarios based on their PQ verification on a day-horizon. Executable load sequences can be automatically combined and scheduled in the system to be adapted to user needs, considering the RE production potential, charge state, and optimal PQ characteristics over the next 24 h. A parametric C++ application software with applied PQ and weather data is free available to allow reproducibility of the results.
Improvements and gaps in the empirical expressions for the fill factor of modern industrial solar cells
2023, Solar Energy Materials and Solar CellsThis study assesses and improves the accuracy of commonly used expressions for the fill factor (FF). Parameters that could affect the accuracy of the revised expressions are investigated. Empirical coefficients of the commonly used analytical expressions are first recalculated using a modified fitting approach. Although the predictions of the revised expressions perfectly match the results of theoretical one-diode model simulations, gaps are observed when compared with actual measurements. The different impacts of unaccounted factors in the expressions are then explored. It is shown that adjusting the ideality factor or considering edge recombination improves the accuracy of the predictions. Moreover, the expressions can slightly overestimate the FF of cells with non-uniform implied open-circuit voltage distribution. As methods to extract electrical parameters from luminescence images continuously improve, the findings of this study can aid in developing techniques for extracting FF from luminescence images of industrial solar cells.
Hardware implementation of an active learning self-organizing neural network to predict the power fluctuation events of a photovoltaic grid-tied system
2022, Microprocessors and MicrosystemsPower fluctuations happens due to highly stochastic solar irradiance nature and this could cause unwanted power quality issues in photovoltaic (PV) system. First predicting the PV output then mitigating the fluctuation by using energy storage system is one of the most effective power fluctuation mitigation method. Incremental learning algorithm such as incremental self-organizing map can be used to actively learn and predict power output. Although significant improvement in terms of mitigated events can be achieved in this kind of system, it still suffers from prediction accuracy due to the highly fluctuating PV profile. In this paper, a time-series unsupervised learning algorithm namely the Time-Series Self-Organizing Incremental Neural Network (TS-SOINN) is proposed to better predict the highly stochastic PV output. It incorporates a novel weighted memory layer to give higher emphasis to recent observation and improves data overlap issues in the conventional self-organizing map algorithm. In the simulation results, the TS-SOINN achieves prediction rate of 93.81% which outperforms the latest unsupervised incremental learning algorithm, M-SOINN by 33.44% and the TS-SOINN mitigated 89.13% power fluctuation events whereas M-SOINN only mitigated 79.62% events. In addition, the TS-SOINN is implemented in Altera Stratix V GS Field Programmable Gate Array (FPGA) board to run real-time prediction. This hardware architecture is able to increase new node and remove inactive nodes throughout the real-time prediction. The hardware TS-SOINN is integrated with mitigation engine to smoothen out power fluctuation events in PV grid-tied system, the experimental results show that the proposed system mitigated 83.33% of power fluctuation events at the grid and the battery state-of-charge maintains within 30%-100%.