Solar Forecasting Research Articles

Algebraic quantification of the contribution of active regions to the Sun’s dipole moment: applications to century-scale polar field estimates and solar cycle forecasting

ABSTRACT The solar cycle is generated by a magnetohydrodynamic dynamo mechanism which involves the induction and recycling of the toroidal and poloidal components of the Sun’s magnetic field. Recent observations indicate that the Babcock–Leighton (BL) mechanism – mediated via the emergence and evolution of tilted bipolar active regions – is the primary contributor to the Sun’s large-scale dipolar field. Surface flux transport models and dynamo models have been employed to simulate this mechanism, which also allows for physics-based solar cycle forecasts. Recently, an alternative analytic method has been proposed to quantify the contribution of individual active regions to the Sun’s dipole moment (DM). Utilizing solar cycle observations spanning a century, here, we test the efficacy of this algebraic approach. Our results demonstrate that the algebraic quantification approach is reasonably successful in estimating DMs at solar minima over the past century – providing a verification of the BL mechanism as the primary contributor to the Sun’s dipole field variations. We highlight that this algebraic methodology serves as an independent approach for estimating DMs at the minima of solar cycles, relying on characteristics of bipolar solar active regions. We also show how this method may be utilized for solar cycle predictions; our estimate of the Sun’s dipole field at the end of cycle 24 using this approach indicates that solar cycle 25 would be a moderately weak cycle, ranging between solar cycle 20 and cycle 24.

Moving beyond the Aerosol Climatology of WRF-Solar: A Case Study over the North China Plain

Abstract Numerical weather prediction (NWP), when accessible, is a crucial input to short-term solar power forecasting. WRF-Solar, the first NWP model specifically designed for solar energy applications, has shown promising predictive capability. Nevertheless, few attempts have been made to investigate its performance under high aerosol loading, which attenuates incoming radiation significantly. The North China Plain is a polluted region due to industrialization, which constitutes a proper testbed for such investigation. In this paper, aerosol direct radiative effect (DRE) on three surface shortwave radiation components (i.e., global, beam, and diffuse) during five heavy pollution episodes is studied within the WRF-Solar framework. Results show that WRF-Solar overestimates instantaneous beam radiation up to 795.3 W m−2 when the aerosol DRE is not considered. Although such overestimation can be partially offset by an underestimation of the diffuse radiation of about 194.5 W m−2, the overestimation of the global radiation still reaches 160.2 W m−2. This undesirable bias can be reduced when WRF-Solar is powered by Copernicus Atmosphere Monitoring Service (CAMS) aerosol forecasts, which then translates to accuracy improvements in photovoltaic (PV) power forecasts. This work also compares the forecast performance of the CAMS-powered WRF-Solar with that of the European Centre for Medium-Range Weather Forecasts model. Under high aerosol loading conditions, the irradiance forecast accuracy generated by WRF-Solar increased by 53.2% and the PV power forecast accuracy increased by 6.8%. Significance Statement Numerical weather prediction (NWP) is the “go-to” approach for achieving high-performance day-ahead solar power forecasting. Integrating time-varying aerosol forecasts into NWP models effectively captures aerosol direct radiation effects, thereby enhancing the accuracy of solar irradiance forecasts in heavily polluted regions. This work not only quantifies the aerosol effects on global, beam, and diffuse irradiance but also reveals the physical mechanisms of irradiance-to-power conversion by constructing a model chain. Using the North China Plain as a testbed, the performance of WRF-Solar on solar power forecasting on five severe pollution days is analyzed. This version of WRF-Solar can outperform the European Centre for Medium-Range Weather Forecasts model, confirming the need for generating high spatial–temporal NWP.

Simulation-based Forecasting for Intraday Power Markets: Modelling Fundamental Drivers for Location, Shape and Scale of the Price Distribution

During the last years, European intraday power markets have gained importance for balancing forecast errors due to the rising volumes of intermittent renewable generation. However, compared to day-ahead markets, the drivers for the intraday price process are still sparsely researched. In this paper, we propose a modelling strategy for the location, shape and scale parameters of the return distribution in intraday markets, based on fundamental variables. We consider wind and solar forecasts and their intraday updates, outages, price information and a novel measure for the shape of the merit-order, derived from spot auction curves as explanatory variables. We validate our modelling by simulating price paths and compare the probabilistic forecasting performance of our model to benchmark models in a forecasting study for the German market. The approach yields significant improvements in the forecasting performance, especially in the tails of the distribution. At the same time, we are able to derive the contribution of the driving variables. We find that, apart from the first lag of the price changes, none of our fundamental variables have explanatory power for the expected value of the intraday returns. This implies weak-form market efficiency as renewable forecast changes and outage information seems to be priced in by the market. We find that the volatility is driven by the merit-order regime, the time to delivery and the closure of cross-border order books. The tail of the distribution is mainly influenced by past price differences and trading activity. Our approach is directly transferable to other continuous intraday markets in Europe.

Artificial Intelligence to Predict Solar Energy Production: Risks and Economic Efficiency

In the context of sustainable development and the increasing shift to non-fossil alternative energy sources, solar energy offers countless advantages for its conversion into electricity. Modern technologies offer great opportunities for the introduction of artificial intelligence in the process of predicting the production of solar energy. However, this topic is still quite unexplored in the international scientific community, which makes this study relevant. The purpose of this study is to analyze the impact of artificial intelligence on solar energy production forecasting. To achieve the purpose of the study, a systematic literature analysis method was applied. As a result of the study, it was possible to establish that artificial intelligence has great potential for its implementation in the process of forecasting solar energy production. The study was able to establish random prediction models and machine learning models based on artificial intelligence for their cost effectiveness and risk in the process of forecasting solar energy production. During the literature review, it became clear that the following four models are the most effective in the work: the RFR, LIME, ELI5 and SHAP. Each model has its own advantages and disadvantages. These are manifested in production management, forecasting with high speed, flexibility, and explanation, reducing the risk of variability. However, the cost-effectiveness of implementing artificial intelligence in the process of forecasting solar energy production has much more economic efficiency than the risk aspects.

Overcoming Language Barrier for Scientific Studies via Unsupervised Literature Learning: Case Study on Solar Cell Materials Prediction

Data‐driven materials and chemical studies have been predominantly confined to English‐language databases, posing challenges for researchers in non‐English‐speaking regions to access and comprehend literature and derive scientific insights. Herein, a machine learning approach designed for information extraction and knowledge acquisition in the materials and chemical science realm from non‐English literature databases, requiring minimal human intervention, is presented. The efficacy of language model through a case study centered on the prediction of solar cell materials using Chinese‐language sources is studied. The unsupervised learning model effectively extracts crucial latent chemical and materials data from non‐English literature resources. Subsequently, the language model successfully identifies existing solar cell materials and forecasts potential candidates from this non‐English corpus. To further validate the suitability of the proposed solar cell material candidates, we conduct ab initio density functional theory calculations to evaluate their structural and optoelectronic properties. The results validate both the efficacy of our language model and the predictability of our approach. This study represents a stride toward comprehensive data‐driven machine learning for materials and chemical predictions, transcending the limitations of English literature. Furthermore, it offers a solution to aid researchers in non‐English‐speaking regions in overcoming language barriers and accessing scientific discoveries.

Evaluation of Various Deep Learning Models for Short-Term Solar Forecasting in the Arctic using a Distributed Sensor Network

Evaluation of Various Deep Learning Models for Short-Term Solar Forecasting in the Arctic using a Distributed Sensor Network

A new evolutionary forest model via incremental tree selection for short-term global solar irradiance forecasting under six various climatic zones

The increasing integration of solar sources into the energy mix presents significant challenges, particularly in short-term energy management. Accurate solar irradiance forecasts can greatly assist solar power plant operators and energy network managers in making informed decisions about energy production and consumption. This paper aims to develop a new accurate forecasting model for short-term global solar irradiance based on an innovative evolutionary forest approach. Our model, baptized EFITS, performs incremental tree selection through appropriate evolutionary operators maintaining a good tradeoff between accuracy and diversity, generating progressively near-optimal decision trees to construct the final evolutionary forest forecaster. This new evolution process also automatically selects near-optimal input parameters, enhancing the overall model accuracy and generalization ability. Six climatically diverse locations in Morocco and three types of inputs (endogenous, exogenous, and hybrid) are used to assess the performance of the proposed. The results demonstrate that our proposed model exhibits excellent performance across all studied sites and horizons. Among all input types, hybrid inputs delivered the best forecasting accuracy across all studied sites and horizons. Notably, the continental climate site (Bni Mellal) achieved the highest accuracy, with nRMSE ranging from 4.94% to 7.54% and nMBE from 0.71% to −0.46% for 1 to 6 h forecasts. Conversely, Ifrane city, characterized by a humid temperate climate, showed the lowest accuracy, with nRMSE ranging from 10.34% to 18.94% and nMBE from 1.21% to −1.54%. Finally, a detailed comparison with benchmarking models (random forest, bagging, gradient boosting, single decision tree, bidirectional long short-term memory network, and scaled persistence models), revealed that our model consistently outperforms them across all tested scenarios, locations, and forecasting horizons.

Economics of physics-based solar forecasting in power system day-ahead scheduling

A high-quality solar power forecasting system that strictly adheres to grid regulations is valuable for system operators to formulate strategies for power system scheduling. Some grid operators, therefore, enact penalty schemes for the forecasts submitted by photovoltaic (PV) plant owners, as a means to fortify truthful and high-quality forecast submissions. From the perspectives of both plant owners and grid operators, this study inquires into the quality-to-value mapping of solar forecasts in the context of power system day-ahead scheduling. A physics-based solar power forecasting method is presented, which consists of two steps. Firstly, ensemble numerical weather prediction (NWP) is summarized into point forecasts. Then irradiance is converted to power via a physical model chain. The results reveal that the two-step physics-based forecasting method has an advantage over a winning method in Global Energy Forecasting Competition 2014 in terms of several accuracy measures. Subsequently, the economics of solar forecasting is quantified through performing day-ahead scheduling on a modified IEEE 30-bus system with PV and battery storage. It is demonstrated that, by respecting the statistical theory on consistency and elicitability when extracting point forecasts from NWP ensembles, both power system operators and PV plant owners can benefit profoundly in terms of cost savings. The former sees fewer needs for reserves, while the latter is less penalized. The data and Python code used to produce the results are also provided to enhance the reproducibility of this work.

Two-Stage Neural Network Optimization for Robust Solar Photovoltaic Forecasting

Neural networks (NNs) have shown outstanding performance in solar photovoltaic (PV) power forecasting due to their ability to effectively learn unstable environmental variables and their complex interactions. However, NNs are limited in their practical industrial application in the energy sector because the optimization of the model structure or hyperparameters is a complex and time-consuming task. This paper proposes a two-stage NN optimization method for robust solar PV power forecasting. First, the solar PV power dataset is divided into training and test sets. In the training set, several NN models with different numbers of hidden layers are constructed, and Optuna is applied to select the optimal hyperparameter values for each model. Next, the optimized NN models for each layer are used to generate estimation and prediction values with fivefold cross-validation on the training and test sets, respectively. Finally, a random forest is used to learn the estimation values, and the prediction values from the test set are used as input to predict the final solar PV power. As a result of experiments in the Incheon area, the proposed method is not only easy to model but also outperforms several forecasting models. As a case in point, with the New-Incheon Sonae dataset—one of three from various Incheon locations—the proposed method achieved an average mean absolute error (MAE) of 149.53 kW and root mean squared error (RMSE) of 202.00 kW. These figures significantly outperform the benchmarks of attention mechanism-based deep learning models, with average scores of 169.87 kW for MAE and 232.55 kW for RMSE, signaling an advance that is expected to make a significant contribution to South Korea’s energy industry.

ASSESSMENT OF REANALYSIS DATA FOR SOLAR PV OUTPUT FORECASTING IN THE PHILIPPINES: CASE OF PANGASINAN, NEGROS OCCIDENTAL, AND DAVAO DEL NORTE

Abstract. The sustainable energy transition in the Philippines requires accurate forecasting of solar PV output to optimize energy efficiency and grid management. While existing studies have emphasized the positive correlation between solar irradiance and PV production, this study aims to explore whether forecasting improves with the inclusion of weather data. This research conducts a comparative analysis between relying solely on solar irradiance against integrating various weather parameters to enhance solar PV output forecasting. The study focuses on three distinct locations (Pangasinan, Negros Occidental, and Davao Del Norte) and employs two models per each site: Model 1 (M1), which relies only on solar irradiance as predictors, and Model 2 (M2), which incorporates solar irradiance and weather parameters. Using Fifth Generation ECMWF Reanalysis (ERA5) Data, Principal Component Analysis (PCA) is conducted on the significant weather parameters. Extreme Gradient Boosting (XGBoost) with 5-fold nested cross-validation is applied for solar PV output forecasting. Models are assessed using Mean Absolute Percentage Error (MAPE) and skill scores. Results showthat while solar irradiance alone suffices for predicting solar PV output in Negros Occidental, incorporating weather parameters improves forecasting accuracy in Davao Del Norte and Pangasinan. This paper recommends caution in generalizing the findings to different regions with varying weather patterns, as the forecasting performance of the models is influenced by data quality, specific location, and prevailing weather conditions.

Day-ahead Numerical Weather Prediction solar irradiance correction using a clustering method based on weather conditions

Accurate solar irradiance forecasts can make solar power forecasts more reliable, which can help the power grid dispatch reasonably. In practice, Numerical Weather Prediction (NWP) is widely applied in solar irradiance forecasts. In this paper, the statistical NWP Global Horizontal Irradiance (GHI) error analysis shows that the characteristics of NWP GHI error vary obviously under different weather conditions. However, existing correction methods are not designed contrapuntally for different weather conditions, resulting in poor correction performance. To solve this problem, a hybrid method is proposed to get day-ahead correction results for NWP GHI. Specifically, the hybrid method consists of three key parts, including Deep Clustering (DC), Variational Mode Decomposition (VMD), and an Encoder–Decoder based Correction model (EDC). DC is used to categorize the historical samples into three clusters, the input of which is the multi-dimensional information series containing observed data and NWP data. After a feature selection by VMD, the correction models are trained on each cluster respectively. The performance of the proposed model is evaluated with a public dataset and an actual field dataset, and the results demonstrate that the accuracy has been effectively improved by the proposed method compared with other models. In addition, we find that adopting VMD is effective in improving the correction accuracy, and the root mean square error is reduced by 5.70% and 9.32% compared with models without it.

Adaptive solar power generation forecasting using enhanced neural network with weather modulation

Solar power generation forecasting plays a vital role in optimizing grid management and stability, particularly in renewable energy-integrated power systems. This research paper presents a comprehensive study on solar power generation forecasting, evaluating traditional and advanced machine learning methods, including ARIMA, Exponential Smoothing, Support Vector Regression, Random Forest, Gradient Boosting, and Physics-based Models. Moreover, we propose an innovative Enhanced Artificial Neural Network (ANN) model, which incorporates Weather Modulation and Leveraging Prior Forecasts to enhance prediction accuracy. The proposed model is evaluated using real-world solar power generation data, and the results demonstrate its superior performance compared to traditional methods and other machine learning approaches. The Enhanced ANN model achieves an impressive Root Mean Square Error (RMSE) of 0.116 and a Mean Absolute Percentage Error (MAPE) of 36.26%. The integration of Weather Modulation allows the model to adapt to changing weather conditions, ensuring reliable forecasts even during adverse scenarios. Leveraging Prior Forecasts enables the model to capture short-term trends, reducing forecasting errors arising from abrupt weather changes. The proposed Enhanced ANN model showcases its potential as a promising tool for precise and reliable solar power generation forecasting, contributing to the efficient integration of solar energy into the power grid and advancing sustainable energy practices.

Improving cross-site generalisability of vision-based solar forecasting models with physics-informed transfer learning

Forecasting solar energy from cloud cover observations is crucial to truly anticipate future changes in power supply. On an intra-hour timescale, ground-level sky cameras located near a solar site offer the most valuable source of information on incoming clouds. In the literature, the analysis of these hyperlocal cloud cover observations for solar modelling is increasingly performed by deep learning algorithms trained and tested on years’ worth of local data. However, this approach is not suitable for industrial applications since solar energy producers cannot wait for years of local data collection to start generating reliable solar forecasts. However, they might own relevant multi-location data collected from other solar sites over time. This study thus explores the capability of such algorithms to generalise beyond their training location in two data scarce conditions: zero-shot learning (i.e. direct application of a trained model to a new location without local fine-tuning) and few-shot learning (i.e. calibration of a pre-trained model based on very limited local data such as a day of observations). Zero-shot learning results show that using local clear-sky models to normalise output variables (e.g. solar irradiance or solar energy production values) facilitates cross-dataset transfer learning. Compared to previous methods, the resulting forecast skill increases by close to 25% in cloudy conditions and by more than 700% in clear-sky conditions. An additional gain is observed when local data collected in overcast weather conditions are used for model calibration via few-shot learning. The corresponding neural networks trained in data scarce conditions achieve comparable performance to expert local models based on years of training data. These promising results shed light on the potential of large-scale and multi-location sky image datasets to improve the generalisation skills of solar forecasting algorithms.

Investigation of Load, Solar and Wind Generation as Target Variables in LSTM Time Series Forecasting, Using Exogenous Weather Variables

Accurately forecasting energy metrics is essential for efficiently managing renewable energy generation. Given the high variability in load and renewable energy power output, this represents a crucial area of research in order to pave the way for increased adoption of low-carbon energy solutions. Whilst the impact of different neural network architectures and algorithmic approaches has been researched extensively, the impact of utilising additional weather variables in forecasts have received far less attention. This article demonstrates that weather variables can have a significant influence on energy forecasting and presents methodologies for using these variables within a long short-term memory (LSTM) architecture to achieve improvements in forecasting accuracy. Moreover, we introduce the use of the seasonal components of the target time series, as exogenous variables, that are also observed to increase accuracy. Load, solar and wind generation time series were forecast one hour ahead using an LSTM architecture. Time series data were collected in five Spanish cities and aggregated for analysis, alongside five exogenous weather variables, also recorded in Spain. A variety of LSTM architectures and hyperparameters were investigated. By tuning exogenous weather variables, a 33% decrease in mean squared error was observed for solar generation forecasting. A 22% decrease in mean absolute squared error (MASE), compared to 24-h ahead forecasts made by the Transmission Service Operator (TSO) in Spain, was also observed for solar generation. Compared to using the target variable in isolation, utilising exogenous weather variables decreased MASE by approximately 10%, 15% and 12% for load, solar and wind generation, respectively. By using the seasonal component of the target variables as an exogenous variable itself, we demonstrated decreases in MASE of 19%, 12% and 8% for load, solar and wind generation, respectively. These results emphasise the significant benefits of incorporating weather and seasonal components into energy-related time series forecasts.

Spatial-temporal characteristics analysis of solar irradiance forecast errors in Europe and North America

Accurate photovoltaic (PV) power forecast is crucial to power systems. Solar irradiance forecast is the fundamentals of PV power forecast. Current researches merely focus on improving the forecast accuracy. There hasn't been comprehensive study on the spatial-temporal characteristics of solar irradiance forecast errors. Thus, this paper analyzed the error characteristics of solar irradiance forecast provided by European Centre for Medium-Range Weather Forecasts (ECMWF), one of the most accurate numerical weather prediction (NWP) products. ECMWF's forecasts were compared with ERA5 solar irradiance reanalysis data to reveal the error distribution characteristics. Four-year (2017–2020) data from the geographical region bounded by 63°N, −126°W, 21°S, and 36°E (covering most parts of Europe and North America) were studied. Experiments show that solar irradiance forecast errors peak at noon of the local time. Furthermore, correlations between solar irradiance forecast errors and other weather variables were also revealed. Experiments show that solar irradiance forecast errors have negative correlation with low cloud cover forecast errors and positive correlation with air temperature forecast errors. The possible reasons for the correlation relationship were also analyzed in detail. This paper systematically reveals the spatial-temporal characteristics of solar irradiance forecast errors and provides a useful guideline for solar PV system operation.

Enhanced Forecasting Accuracy of a Grid-Connected Photovoltaic Power Plant: A Novel Approach Using Hybrid Variational Mode Decomposition and a CNN-LSTM Model

Renewable energies have become pivotal in the global energy landscape. Their adoption is crucial for phasing out fossil fuels and promoting environmentally friendly energy solutions. In recent years, the energy management system (EMS) concept has emerged to manage the power grid. EMS optimizes electric grid operations through advanced metering, automation, and communication technologies. A critical component of EMS is power forecasting, which facilitates precise energy grid scheduling. This research paper introduces a deep learning hybrid model employing convolutional neural network–long short-term memory (CNN-LSTM) for short-term photovoltaic (PV) solar energy forecasting. The proposed method integrates the variational mode decomposition (VMD) algorithm with the CNN-LSTM model to predict PV power output from a solar farm in Boussada, Algeria, spanning 1 January 2019, to 31 December 2020. The performance of the developed model is benchmarked against other deep learning models across various time horizons (15, 30, and 60 min): variational mode decomposition–convolutional neural network (VMD-CNN), variational mode decomposition–long short-term memory (VMD-LSTM), and convolutional neural network–long short-term memory (CNN-LSTM), which provide a comprehensive evaluation. Our findings demonstrate that the developed model outperforms other methods, offering promising results in solar power forecasting. This research contributes to the primary goal of enhancing EMS by providing accurate solar energy forecasts.

Cloud detection method using ground-based sky images based on clear sky library and superpixel local threshold

In the intra-hour time scale, shielding of solar radiation by clouds is the main reason for the fluctuation of photovoltaic, so cloud parameters are important for intra-hour solar irradiance and photovoltaic power forecasting. Extracting cloud regions from cloud images provides a basis for quantifying cloud size and shape. It is difficult to detect clouds in these three cases, such as clouds in circumsolar region, clouds in haze weather and thin clouds at cloud edge. Aiming at such problems, this study firstly establishes a clear sky library based on pixel-level sun positions and haze conditions to deal with the complex changes of sky brightness. Then this study proposes a cloud detection method for ground-based images, which performs two segmentations on the cloud image. The initial segmentation process is a combination method that uses clear sky library method when the sun is visible and uses adaptive threshold method when the sun is occluded. The secondary segmentation process is based on superpixels and local threshold method, which restores some thin clouds that are easily ignored. Finally, for the influence of ghosts, this study summarizes the position and color features of ghosts, and uses different color channel information to deal with them.

Radio Science Experiments during a Cruise Phase to Uranus

The exploration of Uranus, a key archetype for ice giant planets and a gateway to understanding distant exoplanets, is acquiring increasing interest in recent years, especially after the Uranus Orbiter and Probe (UOP) mission has been prioritized in the Planetary Science Decadal Survey 2023–2032. This paper presents the results of numerical simulations aimed at providing experimental constraints on the parameterized post-Newtonian (PPN) parameter γ, a measure of space–time curvature in general relativity (GR), during the cruise phase of a spacecraft travelling to Uranus. Leveraging advanced radio tracking systems akin to those aboard the JUICE and BepiColombo missions, we explore the potential of solar conjunction experiments (SCEs) to refine current measurements of γ by exploiting the spacecraft’s long journey in the outer Solar System. We discuss the anticipated enhancements over previous estimates, underscoring the prospect of detecting violations of GR. Our simulations predict that by using an advanced radio tracking system, it is possible to obtain an improvement in the estimation of γ up to more than an order of magnitude with respect to the latest measurement performed by the Cassini–Huygens mission in 2002, contingent on the calibration capabilities against solar plasma noise. The results reveal that a number of SCEs during the mission can substantially strengthen the validation of GR. In tandem with fundamental physics tests, the use of radio links during SCEs presents a valuable opportunity to dissect the solar corona’s plasma dynamics, contributing to solar physics and space weather forecasting. This paper also enumerates methodologies to analyze electron density, localize plasma features, and deduce solar wind velocity, enriching the scientific yield of the experiments beyond the primary objective of testing GR during the cruise phase of a mission to Uranus.

A Novel Deep Learning-Based Data Analysis Model for Solar Photovoltaic Power Generation and Electrical Consumption Forecasting in the Smart Power Grid

With the installation of solar panels around the world and the permanent fluctuation of climatic factors, it is, therefore, important to provide the necessary energy in the electrical network in order to satisfy the electrical demand at all times for smart grid applications. This study first presents a comprehensive and comparative review of existing deep learning methods used for smart grid applications such as solar photovoltaic (PV) generation forecasting and power consumption forecasting. In this work, electrical consumption forecasting is long term and will consider smart meter data and socioeconomic and demographic data. Photovoltaic power generation forecasting is short term by considering climatic data such as solar irradiance, temperature, and humidity. Moreover, we have proposed a novel hybrid deep learning method based on multilayer perceptron (MLP), long short-term memory (LSTM), and genetic algorithm (GA). We then simulated all the deep learning methods on a climate and electricity consumption dataset for the city of Douala. Electrical consumption data are collected from smart meters installed at consumers in Douala. Climate data are collected at the climate management center in the city of Douala. The results obtained show the outperformance of the proposed optimized method based on deep learning in the both electrical consumption and PV power generation forecasting and its superiority compared to basic methods of deep learning such as support vector machine (SVM), MLP, recurrent neural network (RNN), and random forest algorithm (RFA).

Solar Irradiance Forecasting with Natural Language Processing of Cloud Observations and Interpretation of Results with Modified Shapley Additive Explanations

Forecasting the generation of solar power plants (SPPs) requires taking into account meteorological parameters that influence the difference between the solar irradiance at the top of the atmosphere calculated with high accuracy and the solar irradiance at the tilted plane of the solar panel on the Earth’s surface. One of the key factors is cloudiness, which can be presented not only as a percentage of the sky area covered by clouds but also many additional parameters, such as the type of clouds, the distribution of clouds across atmospheric layers, and their height. The use of machine learning algorithms to forecast the generation of solar power plants requires retrospective data over a long period and formalising the features; however, retrospective data with detailed information about cloudiness are normally recorded in the natural language format. This paper proposes an algorithm for processing such records to convert them into a binary feature vector. Experiments conducted on data from a real solar power plant showed that this algorithm increases the accuracy of short-term solar irradiance forecasts by 5–15%, depending on the quality metric used. At the same time, adding features makes the model less transparent to the user, which is a significant drawback from the point of view of explainable artificial intelligence. Therefore, the paper uses an additive explanation algorithm based on the Shapley vector to interpret the model’s output. It is shown that this approach allows the machine learning model to explain why it generates a particular forecast, which will provide a greater level of trust in intelligent information systems in the power industry.