Generation Forecasting Research Articles

Predicting construction waste in Egyptian residential projects: a robust multiple regression model approach

Effective construction waste (CW) management, mainly concrete, brick, and steel, is a critical challenge due to its significant environmental and economic impacts. This study addresses this challenge by proposing multiple linear regression models to predict waste generation in residential buildings within the Egyptian construction industry, considering the influence of factors such as building design and site management features. Using data from 25 case studies, the models demonstrated high predictive accuracy, with adjusted R² values of 0.877, 0.893, and 0.889 for concrete, bricks, and steel waste, respectively. These R2 values indicate that the models explain approximately 88–89% of the variance in waste generation in residential buildings, highlighting their effectiveness in enhancing resource planning and waste management strategies. The findings suggest that incorporating variables such as total area, design consistency, and site organization significantly improves the accuracy of waste predictions. Although the models show acceptable performance, future research should aim to expand the dataset, incorporate additional variables, and test the models across different types of construction projects to validate further and refine these predictive tools. The models offer valuable insights for enhancing construction practices, minimizing waste, and supporting sustainable development in Egypt’s construction industry. With accurate forecasts of waste generation, the models help project managers and stakeholders to plan CW more effectively, mitigating unnecessary material consumption and reducing environmental impacts. These findings help to adopt sustainable construction practices, such as improved recycling processes and decreased dependence on landfills, to support Egypt’s Vision 2030.

Optimization of Microgrid Dispatching by Integrating Photovoltaic Power Generation Forecast

In order to address the impact of the uncertainty and intermittency of a photovoltaic power generation system on the smooth operation of the power system, a microgrid scheduling model incorporating photovoltaic power generation forecast is proposed in this paper. Firstly, the factors affecting the accuracy of photovoltaic power generation prediction are analyzed by classifying the photovoltaic power generation data using cluster analysis, analyzing its important features using Pearson correlation coefficients, and downscaling the high-dimensional data using PCA. And based on the theories of the sparrow search algorithm, convolutional neural network, and bidirectional long- and short-term memory network, a combined SSA-CNN-BiLSTM prediction model is established, and the attention mechanism is used to improve the prediction accuracy. Secondly, a multi-temporal dispatch optimization model of the microgrid power system, which aims at the economic optimization of the system operation cost and the minimization of the environmental cost, is constructed based on the prediction results. Further, differential evolution is introduced into the QPSO algorithm and the model is solved using this improved quantum particle swarm optimization algorithm. Finally, the feasibility of the photovoltaic power generation forecasting model and the microgrid power system dispatch optimization model, as well as the validity of the solution algorithms, are verified through real case simulation experiments. The results show that the model in this paper has high prediction accuracy. In terms of scheduling strategy, the generation method with the lowest cost is selected to obtain an effective way to interact with the main grid and realize the stable and economically optimized scheduling of the microgrid system.

A high-resolution three-year dataset supporting rooftop photovoltaics (PV) generation analytics

This paper presents an open-source dataset intended to enhance the analysis and optimization of photovoltaic (PV) power generation in urban environments, serving as a valuable resource for various applications in solar energy research and development. The dataset comprises measured PV power generation data and corresponding on-site weather data gathered from 60 grid-connected rooftop PV stations in Hong Kong over a three-year period (2021-2023). The PV power generation data was collected at 5-minute intervals at the inverter-level. The meteorological data was collected at 1-minute intervals from an on-site weather station. The metadata was represented using the Brick schema, which simplifies data comprehension and the development of smart analytics applications. This paper provides a detailed description on the site specifications, data collection method, data records, and data validation. This dataset can be used in various applications - PV generation benchmarking, PV degradation analysis, PV fault detection, solar radiation and PV power generation forecasting, and the simulation and design of PV systems.

Enhanced Deep Learning Based Forecasting of Solar Photovoltaic Generation for Critical Weather Conditions

Enhanced Deep Learning Based Forecasting of Solar Photovoltaic Generation for Critical Weather Conditions

Hybridizing Machine Learning Algorithms With Numerical Models for Accurate Wind Power Forecasting

ABSTRACTAn accurate prediction of wind power generation is crucial for optimizing the integration of wind energy into the power grid, ensuring energy reliability. This research focuses on enhancing the accuracy of wind power generation forecasts by combining data from mesoscale and reanalysis models with Machine Learning (ML) approaches. We utilized WRF forecast data alongside ERA5 reanalysis data to estimate wind power generation for a wind farm located at Valladolid, Spain. The study evaluated the performance of ML models based on WRF and ERA5 data individually, as well as a combined model using inputs from both datasets. The hybrid model combining WRF and ERA5 data with ML resulted in a 15% improvement in root mean square error (RMSE) and a 10% increase in compared with standalone models, providing a more reliable 1‐h forecast of wind power generation. Additionally, the availability of data over time was addressed: WRF provides the advantage of projecting data into the future, whereas ERA5 offers retrospective data.

TBFL: blockchain-enabled trusted byzantine-robust federated learning framework for photovoltaic power generation forecasting

Abstract Precise forecasting of photovoltaic (PV) power generation upholds flexibility and reliability within the power grid. Due to the data security dilemma of previous forecasting methods, federated learning (FL) has been widely studied for its ability to train models without sharing training data. However, the incorrect behavior from untrusted devices and servers in traditional FL frameworks can undermine the integrity of the global model, precipitating inaccurate power generation forecasting. Therefore, we propose a blockchain-enabled trusted Byzantine-robust FL framework, called TBFL, designed for decentralized and privacy-preserving PV power generation forecasting. Specifically, this framework features a trusted supervision mechanism, which can effectively eliminate malicious gradients to achieve a high-quality model. In addition, a multilevel differential privacy scheme is designed to strike a balance between privacy protection and model accuracy. Finally, a model clipping algorithm based on neuronal similarity is implemented to optimize both the duration and consumption associated with local device training. Comprehensive experimental outcomes demonstrate that the framework TBFL can successfully improve robustness, and achieve similar efficiency as FedAvg while maintaining a high forecasting accuracy.

Causal Mechanism-Enabled Zero-Label Learning for Power Generation Forecasting of Newly-Built PV Sites

Causal Mechanism-Enabled Zero-Label Learning for Power Generation Forecasting of Newly-Built PV Sites

Estimation of Hydroelectric Power Generation Forecasting and Analysis of Climate Factors with Deep Learning Methods: A Case Study in Yozgat Province in Turkey

Hydroelectric power is a significant renewable energy source for the development of countries. However, climatic data can impact power generation in hydroelectric power plants. Hydroelectric power forecasting is conducted in this study using Long Short-Term Memory (LSTM), Support Vector Regression (SVR), and hybrid LSTM-SVR models based on climatic data. The dataset consists of climate data from the Yozgat Meteorology Directorate in Turkey from 2007 to 2021 and power data obtained from the Süreyyabey Hydroelectric Power Plant in Yozgat. The correlation coefficient examines the relationship between climate data and monthly hydroelectric power generation. The hyper-parameters of the models are adjusted using the Bayesian Optimization (BO) method. The performance of monthly hydroelectric power prediction models is assessed using metrics such as correlation coefficient (R), root mean square error (RMSE), and mean absolute error (MAE). When trained using 11 and 12 climate parameters, the SVR model exhibits an R-value close to 1, and MAE and RMSE values close to 0 are observed. Additionally, regarding training time, the SVR model achieves accurate predictions with the shortest duration and the least error compared to other models.

Validation of the phenological forecast model "RIMpro-Cydia 3" in the region of Plovdiv (Central South Bulgaria)

The study aimed to validate the forecast model "RIMpro-Cydia 3" for the region of the city of Plovdiv and through it to signalize the most appropriate periods for control of the codling moth Cydia pomonella L. (Lepidoptera: Tortricidae) at the conditions of Bulgaria. The observations of the phenology of the pest were conducted in the period 2020 – 2022 in apple orchards in the region of Plovdiv. The model used the data from the automatic weather station, installed in the Centre of Integrated Disease Management at the Experimental Field of the Agricultural University, Plovdiv. Comparing the prediction of the model with the observed values in the region of Plovdiv, the codling moth developеа two full and partial third generations per year, which overlap. Their population dynamics and density depend both on the weather conditions and on the applied plant protection measures. The second generation of the pest had a higher density and causes greater losses to the fruit crop. In this region “RIMpro-Cydia 3” model predicted correctly the development and population density of the first generation of the codling moth, while for the second generation, the model showed significant deviations in the prediction of the population density as well as the exact development periods. The farmers have to use the model data only for the forecast of the first generation of the Codling moth. The measures for efficient control of the second generation should be signalized by direct observations in the orchards. Keywords: codling moth, Cydia pomonella, phenological development, forecast model, RIMpro-Cydia 3

A Study on Clustering Method for Solar Power Generation Forecasting in Renewable Energy Monitoring System

A Study on Clustering Method for Solar Power Generation Forecasting in Renewable Energy Monitoring System

Improving Model Chain Approaches for Probabilistic Solar Energy Forecasting through Post-processing and Machine Learning

Weather forecasts from numerical weather prediction models play a central role in solar energy forecasting, where a cascade of physics-based models is used in a model chain approach to convert forecasts of solar irradiance to solar power production. Ensemble simulations from such weather models aim to quantify uncertainty in the future development of the weather, and can be used to propagate this uncertainty through the model chain to generate probabilistic solar energy predictions. However, ensemble prediction systems are known to exhibit systematic errors, and thus require post-processing to obtain accurate and reliable probabilistic forecasts. The overarching aim of our study is to systematically evaluate different strategies to apply post-processing in model chain approaches with a specific focus on solar energy: not applying any post-processing at all; post-processing only the irradiance predictions before the conversion; post-processing only the solar power predictions obtained from the model chain; or applying post-processing in both steps. In a case study based on a benchmark dataset for the Jacumba solar plant in the U.S., we develop statistical and machine learning methods for post-processing ensemble predictions of global horizontal irradiance (GHI) and solar power generation. Further, we propose a neural-network-based model for direct solar power forecasting that bypasses the model chain. Our results indicate that postprocessing substantially improves the solar power generation forecasts, in particular when post-processing is applied to the power predictions. The machine learning methods for post-processing slightly outperform the statistical methods, and the direct forecasting approach performs comparably to the post-processing strategies.

Rules-Based Energy Management System for an EV Charging Station Nanogrid: A Stochastic Analysis

The article presents the development of a Rules-Based Energy Management System for a nanogrid that serves an electric vehicle charging station. This nanogrid is composed of photovoltaic generation, a wind turbine, a battery energy storage system, and a fast electric vehicle charger. The objective is to prioritize the use of renewable energy sources, reducing costs and promoting energy efficiency. The methodology includes forecasting models based on an Artificial Neural Network for photovoltaic generation, a parametric estimation for wind generation, and a Monte Carlo simulation to predict the energy consumption of electric vehicles. The developed algorithm makes decisions every 15 min, considering variables such as energy tariff, battery state of charge, renewable generation forecast, and energy consumption forecast. The results showed that the system adequately balances energy generation, consumption, and storage, even under forecasting uncertainties. The use of the Monte Carlo simulation was crucial for evaluating the financial impacts of forecast errors, enabling robust decision-making. This energy management system proved to be effective and sustainable for nanogrids dedicated to electric vehicle charging, with the potential to reduce operational costs and increase energy reliability and the use of renewable energy sources.

Short-Term Energy Generation Forecasts at a Wind Farm—A Multi-Variant Comparison of the Effectiveness and Performance of Various Gradient-Boosted Decision Tree Models

High-quality short-term forecasts of wind farm generation are crucial for the dynamically developing renewable energy generation sector. This article addresses the selection of appropriate gradient-boosted decision tree models (GBDT) for forecasting wind farm energy generation with a 10-min time horizon. In most forecasting studies, authors utilize a single gradient-boosted decision tree model and compare its performance with other machine learning (ML) techniques and sometimes with a naive baseline model. This paper proposes a comprehensive comparison of all gradient-boosted decision tree models (GBDTs, eXtreme Gradient Boosting (XGBoost), Light Gradient-Boosting Machine (LightGBM), and Categorical Boosting (CatBoost)) used for forecasting. The objective is to evaluate each model in terms of forecasting accuracy for wind farm energy generation (forecasting error) and computational time during model training. Computational time is a critical factor due to the necessity of testing numerous models with varying hyperparameters to identify the optimal settings that minimize forecasting error. Forecast quality using default hyperparameters is used here as a reference. The research also seeks to determine the most effective sets of input variables for the predictive models. The article concludes with findings and recommendations regarding the preferred GBDT models. Among the four tested models, the oldest GBDT model demonstrated a significantly longer training time, which should be considered a major drawback of this implementation of gradient-boosted decision trees. In terms of model quality testing, the lowest nRMSE error was achieved by the oldest model—GBDT in its tuned version (with the best hyperparameter values obtained from exploring 40,000 combinations).

Enhanced Solar Power Prediction Using Attention-Based DiPLS-BiLSTM Model

The data for solar power generation contain a huge amount of data with a large number of features which are difficult to extract effectively. It is important for the grid management and operational efficiency of the solar farm to accurately predict the solar power. The existing prediction models utilize historical data but often fail to capture critical latent features. This limitation leads to overlooked complex dependencies or temporal relationships, reducing prediction accuracy, especially in load and generation forecasting. An attention-based dynamic inner partial least squares (DiPLS) model and a bidirectional long short-term memory (BiLSTM) model were used for solar power prediction. First, DiPLS is used to dynamically extract features, and then, an attention process is used to predict the importance of these features. Finally, the raw sets are input to the BiLSTM model to make predictions of solar power in the future. The proposed method improves prediction accuracy, achieving an R2 value of 0.965 for training and 0.961 for testing, compared to conventional models. Additionally, the method demonstrated lower root mean squared error (RMSE), indicating enhanced stability and accuracy for solar power forecasting.

Performance Comparison of Machine Learning Algorithms for Wind Energy Forecasting in the Coastal Region of Kerala

This paper presents the details predicting wind energy output with machine learning models. Accurate forecasts of wind power's future are essential to the feasibility of major renewable energy projects. Making precise forecasts of wind power generation requires accounting for changes in weather patterns over time. This is also essential for issuing early warnings and implementing risk-reduction measures. In this study, prediction models for wind energy are developed using wind data collected at coastal regions in Kerala. Accurate forecasting of wind power generation is necessary to balance supply and demand in the smart grid. In the present investigation, an extensive analysis of long-term wind power forecasting was undertaken utilizing daily wind speed data, employing five distinct machine learning algorithms like XG Boost, LASSO, Gradient Boosting, Random Forest, and Bayesian Ridge Regression.

Ground-Based Cloud Image Segmentation Method Based on Improved U-Net

Cloud image segmentation is a technique that divides images captured by meteorological satellites or ground-based observations into different regions or categories. By extracting the distribution, shape, and dynamic features of clouds, it provides precise data support for the meteorological and environmental fields, significantly influencing photovoltaic (PV) power generation forecasting, astronomical telescope observatory site selection, and weather forecasting. A ground-based cloud image segmentation model based on an improved U-Net is proposed, which adopts an overall encoder–decoder structure. In the encoder phase, this paper constructs a dilated convolution–atrous spatial pyramid pooling (ASPP)–dilated convolution structure to enhance early cloud feature extraction. Dilated convolution is a novel type of convolution that expands the receptive field by inserting holes into standard convolution, thereby capturing a larger range of contextual information. ASPP maintains high resolution while paying attention to both local details and global structures of the image. In the decoder stage, the bicubic interpolation method is used for up-sampling to restore the feature map resolution and improve the clarity of the segmented image. The bicubic interpolation method refers to the use of cubic polynomial functions to interpolate the pixel values of the input image. In addition, this paper designs a novel skip connection layer structure between the encoder and decoder, composed of a depthwise separable path (DS path) and an improved channel spatial attention module (Im-CSAM) connected in sequence. The DS path combines depthwise separable convolutions and residual structures to facilitate information exchange between high-level and low-level features. The Im-CSAM is a modular attention mechanism that focuses on important cloud features in spatial and channel dimensions to enhance segmentation accuracy. Experiments show that compared to the traditional U-Net, the accuracy, precision, and MIoU of this model improved by 2.2%, 4.1%, and 5.0%, respectively, in the SWINySEG dataset, and by 3.2%, 3.6%, and 5.8%, respectively, in the TCDD dataset, proving that the improved method has a better generalization ability and segmentation performance.

A hybrid deep learning framework for short-term load forecasting with improved data cleansing and preprocessing techniques

A hybrid deep learning framework for short-term load forecasting with improved data cleansing and preprocessing techniques

Tree-Based Forecasting of Day-Ahead Solar Power Generation from Granular Meteorological Features

Tree-Based Forecasting of Day-Ahead Solar Power Generation from Granular Meteorological Features

Solar Power Generation Forecasting in Smart Cities and Explanation Based on Explainable AI

Solar Power Generation Forecasting in Smart Cities and Explanation Based on Explainable AI

Discrete control model Q-learning for an energy storage system with a hydrogen unit of an autonomous hybrid power plant of a railway substation

Discrete control model Q-learning for an energy storage system with a hydrogen unit of an autonomous hybrid power plant of a railway substation