Solar radiation forecasting using MARS, CART, M5, and random forest model: A case study for India

Prediction of daily global solar radiation in different climatic conditions using metaheuristic search algorithms: a case study from Türkiye

Today's many giant sectors including energy, industry, tourism, and agriculture should closely track the variation trends of solar radiation to take more benefit from the sun. However, the scarcity of solar radiation measuring stations represents a significant obstacle. This has prompted research into the estimation of global solar radiation (GSR) for various regions using existing climatic and atmospheric parameters. While prediction methods cannot supplant the precision of direct measurements, they are invaluable for studying and utilizing solar energy on a global scale. From this point of view, this paper has focused on predicting daily GSR data in three provinces (Afyonkarahisar, Rize, and Ağrı) which exhibit disparate solar radiation distributions in Türkiye. In this context, Gradient-Based Optimizer (GBO), Harris Hawks Optimization (HHO), Barnacles Mating Optimizer (BMO), Sine Cosine Algorithm (SCA), and Henry Gas Solubility Optimization (HGSO) have been employed to model the daily GSR data. The algorithms were calibrated with daily historical data of five input variables including sunshine duration, actual pressure, moisture, wind speed, and ambient temperature between 2010 and 2017 years. Then, they were tested with daily data for the 2018 year. In the study, a series of statistical metrics (R2, MABE, RMSE, and MBE) were employed to elucidate the algorithm that predicts solar radiation data with higher accuracy. The prediction results demonstrated that all algorithms achieved the highest R2 value in Rize province. It has been found that SCA (MABE of 0.7023 MJ/m2, RMSE of 0.9121 MJ/m2, and MBE of 0.2430 MJ/m2) for Afyonkarahisar province and GBO (RMSE of 0.8432 MJ/m2, MABE of 0.6703 MJ/m2, and R2 of 0.8810) for Ağrı province are the most effective algorithms for estimating GSR data. The findings indicate that each of the metaheuristic algorithms tested in this paper has the potential to predict daily GSR data within a satisfactory error range. However, the GBO and SCA algorithms provided the most accurate predictions of daily GSR data.

An ensemble deep learning approach for predicting cocoa yield

One important aspect of agriculture is crop yield prediction. This aspect allows decision-makers and farmers to make adequate planning and policies. Before now, various statistical models have been used for crop yield prediction but this approach experienced some hiccups such as time wastage, inaccurate prediction, and difficulties in model usage. Recently, a new trend of deep learning and machine learning are now adopted for crop yield prediction. Deep learning can extract patterns from a large volume of the dataset, thus, they are suitable for prediction. The research work aims to propose an efficient deep-learning technique in the field of cocoa yield prediction. This research presents a deep learning approach for cocoa yield prediction using a Convolutional Neural Network and Recurrent Neural Network (CNN-RNN) with Long Short Term Memory (LSTM). The ensemble approach was adopted because of the nature of the dataset used. Two different sets of the dataset were used, namely; the climatic dataset and the cocoa yield dataset. CNN-RNN with LSTM has some salient features, where CNN was used to handle the climatic dataset, and RNN was employed to handle the cocoa yield prediction in southwest Nigeria. Two major problems generated by the CNN-RNN model are vanishing and exploding gradients and this was handled by LSTM. The proposed model was benchmarked with other machine learning algorithms based on Mean Absolute Error (MAE), Mean Square Error (MSE), Root Mean Square Error (RMSE), and Mean Absolute Percentage Error (MAPE). CNN-RNN with LSTM gave the least mean of absolute error as compared to the other machine learning algorithms which shows the efficiency of the model.

CyL-GHI: Global Horizontal Irradiance Dataset Containing 18 Years of Refined Data at 30-Min Granularity from 37 Stations Located in Castile and León (Spain)

Accurate solar forecasting lately relies on advances in the field of artificial intelligence and on the availability of databases with large amounts of information on meteorological variables. In this paper, we present the methodology applied to introduce a large-scale, public, and solar irradiance dataset, CyL-GHI, containing refined data from 37 stations found within the Spanish region of Castile and León (Spanish: Castilla y León, or CyL). In addition to the data cleaning steps, the procedure also features steps that enable the addition of meteorological and geographical variables that complement the value of the initial data. The proposed dataset, resulting from applying the processing methodology, is delivered both in raw format and with the quality processing applied, and continuously covers 18 years (the period from 1 January 2002 to 31 December 2019), with a temporal resolution of 30 min. CyL-GHI can result in great importance in studies focused on the spatial-temporal characteristics of solar irradiance data, due to the geographical information considered that enables a regional analysis of the phenomena (the 37 stations cover a land area larger than 94,226 km2). Afterwards, three popular artificial intelligence algorithms were optimised and tested on CyL-GHI, their performance values being offered as baselines to compare other forecasting implementations. Furthermore, the ERA5 values corresponding to the studied area were analysed and compared with performance values delivered by the trained models. The inclusion of previous observations of neighbours as input to an optimised Random Forest model (applying a spatio-temporal approach) improved the predictive capability of the machine learning models by almost 3%.

A state of art review on estimation of solar radiation with various models

Solar radiation is free, and very useful input for most sectors such as heat, health, tourism, agriculture, and energy production, and it plays a critical role in the sustainability of biological, and chemical processes in nature. In this framework, the knowledge of solar radiation data or estimating it as accurately as possible is vital to get the maximum benefit from the sun. From this point of view, many sectors have revised their future investments/plans to enhance their profit margins for sustainable development according to the knowledge/estimation of solar radiation. This case has noteworthy attracted the attention of researchers for the estimation of solar radiation with low errors. Accordingly, it is noticed that various types of models have been continuously developed in the literature. The present review paper has mainly centered on the solar radiation works estimated by the empirical models, time series, artificial intelligence algorithms, and hybrid models. In general, these models have needed the atmospheric, geographic, climatic, and historical solar radiation data of a given region for the estimation of solar radiation. It is seen from the literature review that each model has its advantages and disadvantages in the estimation of solar radiation, and a model that gives the best results for one region may give the worst results for the other region. Furthermore, it is noticed that an input parameter that strongly improves the performance success of the models for a region may worsen the performance success of another region. In this direction, the estimation of solar radiation has been separately detailed in terms of empirical models, time series, artificial intelligence algorithms, and hybrid algorithms. Accordingly, the research gaps, challenges, and future directions for the estimation of solar radiation have been drawn in the present study. In the results, it is well-observed that the hybrid models have exhibited more accurate and reliable results in most studies due to their ability to merge between different models for the benefit of the advantages of each model, but the empirical models have come to the fore in terms of ease of use, and low computational costs.

A Comprehensive Review on Ensemble Solar Power Forecasting Algorithms

With increasing demand for energy, the penetration of alternative sources such as renewable energy in power grids has increased. Solar energy is one of the most common and well-known sources of energy in existing networks. But because of its non-stationary and non-linear characteristics, it needs to predict solar irradiance to provide more reliable Photovoltaic (PV) plants and manage the power of supply and demand. Although there are various methods to predict the solar irradiance. This paper gives the overview of recent studies with focus on solar irradiance forecasting with ensemble methods which are divided into two main categories: competitive and cooperative ensemble forecasting. In addition, parameter diversity and data diversity are considered as competitive ensemble forecasting and also preprocessing and post-processing are as cooperative ensemble forecasting. All these ensemble forecasting methods are investigated in this study. In the end, the conclusion has been drawn and the recommendations for future studies have been discussed.

Trend analysis and forecasting of streamflow using random forest in the Punarbhaba River basin

Streamflow rate changes due to damming are hydro-ecologically sensitive in present and future times. Very less studies have done an investigation of the damming effect on the streamflow along with future forecasting, which can be the solution for the existing problems. Therefore, this study aims to use the Pettitt test as well as standard normal homogeneity test (SNHT) to discover trends in streamflow with the future situation in the Punarbhaba River in Indo-Bangladesh from 1978 to 2017. Trend was spotted using Mann-Kendall test, Spearman's rank correlation approach, innovative trend analysis, and a linear regression model. The current work additionally uses advanced machine learning techniques like random forest (RF) to estimate flow regimes using historical time series data. 1992 appears to be a yard mark in this continuum of time series datasets, indicating a significant transformation in the streamflow regime. The MK test as well as Spearman's rho was used to find a significant negative trend for the average (-0.57), maximum (-0.62), and minimum (-0.48) flow regimes. The consistency of the flow regime has been losing consistency, and the variability of flow regime has increased from 2.1 to 6.7% of the average water level, 1.5 to 6.5% of the maximum streamflow, and 3.1 to 5.8% of the minimum streamflow in the post-change point phase. The forecast trend using random forest for streamflow up to 2030 are negative for all four seasons with a flow volume likely to be reduced by 0.67% to-5.23%. Annual and monthly streamflows revealed very negative tendencies, according to the conclusions of unique trend analysis. Flow declination of this magnitude impacts downstream habitat and environment. According to future estimates, the seasonal flow will decrease. Furthermore, the outcome of this research will give a wealth of data for river management and other places with comparable environment.

Optimized Random Forest for Solar Radiation Prediction Using Sunshine Hours

Knowing exactly how much solar radiation reaches a particular area is helpful when planning solar energy installations. In recent years the use of renewable energies, especially those related to photovoltaic systems, has had an impressive up-tendency. Therefore, mechanisms that allow us to predict solar radiation are essential. This work aims to present results for predicting solar radiation using optimization with the Random Forest (RF) algorithm. Moreover, it compares the obtained results with other machine learning models. The conducted analysis is performed in Queretaro, Mexico, which has both direct solar radiation and suitable weather conditions more than three quarters of the year. The results show an effective improvement when optimizing the hyperparameters of the RF and Adaboost models, with an improvement of 95.98% accuracy compared to conventional methods such as linear regression, with 54.19%, or recurrent networks, with 53.96%, without increasing the computational time and performance requirements to obtain the prediction. The analysis was successfully repeated in two different scenarios for periods in 2020 and 2021 in Juriquilla. The developed method provides robust performance with similar results, confirming the validity and effectiveness of our approach.

Comprehensive assessment, review, and comparison of AI models for solar irradiance prediction based on different time/estimation intervals

Solar energy-based technologies have developed rapidly in recent years, however, the inability to appropriately estimate solar energy resources is still a major drawback for these technologies. In this study, eight different artificial intelligence (AI) models namely; convolutional neural network (CNN), artificial neural network (ANN), long short-term memory recurrent model (LSTM), eXtreme gradient boost algorithm (XG Boost), multiple linear regression (MLR), polynomial regression (PLR), decision tree regression (DTR), and random forest regression (RFR) are designed and compared for solar irradiance prediction. Additionally, two hybrid deep neural network models (ANN-CNN and CNN-LSTM-ANN) are developed in this study for the same task. This study is novel as each of the AI models developed was used to estimate solar irradiance considering different timesteps (hourly, every minute, and daily average). Also, different solar irradiance datasets (from six countries in Africa) measured with various instruments were used to train/test the AI models. With the aim to check if there is a universal AI model for solar irradiance estimation in developing countries, the results of this study show that various AI models are suitable for different solar irradiance estimation tasks. However, XG boost has a consistently high performance for all the case studies and is the best model for 10 of the 13 case studies considered in this paper. The result of this study also shows that the prediction of hourly solar irradiance is more accurate for the models when compared to daily average and minutes timestep. The specific performance of each model for all the case studies is explicated in the paper.

Machine Learning and Deep Learning in Energy Systems: A Review

With population increases and a vital need for energy, energy systems play an important and decisive role in all of the sectors of society. To accelerate the process and improve the methods of responding to this increase in energy demand, the use of models and algorithms based on artificial intelligence has become common and mandatory. In the present study, a comprehensive and detailed study has been conducted on the methods and applications of Machine Learning (ML) and Deep Learning (DL), which are the newest and most practical models based on Artificial Intelligence (AI) for use in energy systems. It should be noted that due to the development of DL algorithms, which are usually more accurate and less error, the use of these algorithms increases the ability of the model to solve complex problems in this field. In this article, we have tried to examine DL algorithms that are very powerful in problem solving but have received less attention in other studies, such as RNN, ANFIS, RBN, DBN, WNN, and so on. This research uses knowledge discovery in research databases to understand ML and DL applications in energy systems’ current status and future. Subsequently, the critical areas and research gaps are identified. In addition, this study covers the most common and efficient applications used in this field; optimization, forecasting, fault detection, and other applications of energy systems are investigated. Attempts have also been made to cover most of the algorithms and their evaluation metrics, including not only algorithms that are more important, but also newer ones that have received less attention.

Solar Irradiance Forecasting Using Dynamic Ensemble Selection

Solar irradiance forecasting has been an essential topic in renewable energy generation. Forecasting is an important task because it can improve the planning and operation of photovoltaic systems, resulting in economic advantages. Traditionally, single models are employed in this task. However, issues regarding the selection of an inappropriate model, misspecification, or the presence of random fluctuations in the solar irradiance series can result in this approach underperforming. This paper proposes a heterogeneous ensemble dynamic selection model, named HetDS, to forecast solar irradiance. For each unseen test pattern, HetDS chooses the most suitable forecasting model based on a pool of seven well-known literature methods: ARIMA, support vector regression (SVR), multilayer perceptron neural network (MLP), extreme learning machine (ELM), deep belief network (DBN), random forest (RF), and gradient boosting (GB). The experimental evaluation was performed with four data sets of hourly solar irradiance measurements in Brazil. The proposed model attained an overall accuracy that is superior to the single models in terms of five well-known error metrics.

Optimization algorithms as training approach with hybrid deep learning methods to develop an ultraviolet index forecasting model

The solar ultraviolet index (UVI) is a key public health indicator to mitigate the ultraviolet-exposure related diseases. This study aimed to develop and compare the performances of different hybridised deep learning approaches with a convolutional neural network and long short-term memory referred to as CLSTM to forecast the daily UVI of Perth station, Western Australia. A complete ensemble empirical mode decomposition with adaptive noise (CEEMDAN) is incorporated coupled with four feature selection algorithms (i.e., genetic algorithm (GA), ant colony optimization (ACO), particle swarm optimization (PSO), and differential evolution (DEV)) to understand the diverse combinations of the predictor variables acquired from three distinct datasets (i.e., satellite data, ground-based SILO data, and synoptic mode climate indices). The CEEMDAN-CLSTM model coupled with GA appeared to be an accurate forecasting system in capturing the UVI. Compared to the counterpart benchmark models, the results demonstrated the excellent forecasting capability (i.e., low error and high efficiency) of the recommended hybrid CEEMDAN-CLSTM model in apprehending the complex and non-linear relationships between predictor variables and the daily UVI. The study inference can considerably enhance real-time exposure advice for the public and help mitigate the potential for solar UV-exposure-related diseases such as melanoma.

Research on the Gas Emission Quantity Prediction Model of Improved Artificial Bee Colony Algorithm and Weighted Least Squares Support Vector Machine (IABC-WLSSVM)

In order to further accurately predict gas emission of working face, this paper proposes a prediction model of gas emission of working face based on the combination of improved artificial bee colony algorithm and weighted least squares support vector machine (IABC-WLSSAVM). The research steps are as follows: Firstly, in order to obtain the sparse solution of LSSVM, a more reliable prediction model is realized by weighting the error value. Secondly, the chaotic sequence is introduced into the artificial bee colony algorithm to find a better initial honey source, which increases the diversity of the population, and combines the Levy flight to update the search step to avoid falling into the trap of local optimum. At the same time, the improved artificial bee colony algorithm is used to optimize the kernel width σ and regularization parameter λ of WLSSVM, which improves the prediction accuracy and convergence rate of WLSSVM. Finally, the quantitative analysis model of WLSSVM is reconstructed by using the optimized parameters, and the nine parameters of buried depth of coal seam, gas content of coal seam, coal thickness, interlayer lithology, production rate of working face, length of working face, inclination of coal seam, gas content of adjacent layer, and thickness of adjacent layer are used as the main influencing factors. After normalization, the nonlinear prediction model of gas emission is established. The simulation results based on the three indicators of determination coefficient, root mean square error, and average relative variance show that the IABC-WLSSVM prediction model proposed in this paper can not only overcome the local optimization to obtain the global optimal solution but also has faster convergence speed and higher prediction accuracy. This prediction model has obvious advantages compared with the other three improved prediction models in terms of fitting, accuracy, and generalization ability, which can provide a reliable theoretical basis for the prediction of gas emission in coal mining face under complex factors and propose a new idea for the application of artificial intelligence in the construction of intelligent mines. At the same time, the prediction model can also be applied to other fields.

Modelling and Prediction of Monthly Global Irradiation Using Different Prediction Models

Different prediction models (multiple linear regression, vector support machines, artificial neural networks and random forests) are applied to model the monthly global irradiation (MGI) from different input variables (latitude, longitude and altitude of meteorological station, month, average temperatures, among others) of different areas of Galicia (Spain). The models were trained, validated and queried using data from three stations, and each best model was checked in two independent stations. The results obtained confirmed that the best methodology is the ANN model which presents the lowest RMSE value in the validation and querying phases 1226 kJ/(m2∙day) and 1136 kJ/(m2∙day), respectively, and predict conveniently for independent stations, 2013 kJ/(m2∙day) and 2094 kJ/(m2∙day), respectively. Given the good results obtained, it is convenient to continue with the design of artificial neural networks applied to the analysis of monthly global irradiation.

Application of M5 model tree optimized with Excel Solver Platform for water quality parameter estimation

The high cost and time for determining water quality parameters justify the importance of application of mathematical models in discovering connection among them. This paper presents a data mining technique and its improved version in estimating water quality parameters. For this purpose, the surface and ground water quality data from Hamedan (Iran) between 2006 and 2015 were analyzed using M5 model tree and its modified version optimized with Excel Solver Platform (ESP). The values of electrical conductivity (EC), total dissolved solids (TDS), sodium adsorption ratio (SAR), and total hardness (TH) were considered as target variables, whereas pH, concentrations of sodium (Na), chlorine (Cl), bicarbonate (HCO₃), sulfate (SO₄), magnesium (Mg), calcium (Ca), and potassium (K) were as inputs. The results showed that in both the sources, pH was the least influential parameter on EC, TDS, SAR, and TH. It was found that among the objective parameters, the accuracy of models in estimating TH was higher than the other parameters, whereas SAR was a complex variable. The comparison of performances of the M5 and the M5-ESP models illustrated that the application of the ESP significantly decreased the normal root mean error (NRMSE) of the M5 model; the mean NRMSEs were decreased by 18.95% and 20.29% in estimating groundwater and surface water quality parameters, respectively. Moreover, ability of both the M5 and the M5-ESP models in computing objective parameters of the groundwater was found to be better than the surface water.

Mapping wind erosion hazard with regression-based machine learning algorithms

Land susceptibility to wind erosion hazard in Isfahan province, Iran, was mapped by testing 16 advanced regression-based machine learning methods: Robust linear regression (RLR), Cforest, Non-convex penalized quantile regression (NCPQR), Neural network with feature extraction (NNFE), Monotone multi-layer perception neural network (MMLPNN), Ridge regression (RR), Boosting generalized linear model (BGLM), Negative binomial generalized linear model (NBGLM), Boosting generalized additive model (BGAM), Spline generalized additive model (SGAM), Spike and slab regression (SSR), Stochastic gradient boosting (SGB), support vector machine (SVM), Relevance vector machine (RVM) and the Cubist and Adaptive network-based fuzzy inference system (ANFIS). Thirteen factors controlling wind erosion were mapped, and multicollinearity among these factors was quantified using the tolerance coefficient (TC) and variance inflation factor (VIF). Model performance was assessed by RMSE, MAE, MBE, and a Taylor diagram using both training and validation datasets. The result showed that five models (MMLPNN, SGAM, Cforest, BGAM and SGB) are capable of delivering a high prediction accuracy for land susceptibility to wind erosion hazard. DEM, precipitation, and vegetation (NDVI) are the most critical factors controlling wind erosion in the study area. Overall, regression-based machine learning models are efficient techniques for mapping land susceptibility to wind erosion hazards.