To see the other types of publications on this topic, follow the link: Power predictions.
Journal articles on the topic 'Power predictions'
Create a spot-on reference in APA, MLA, Chicago, Harvard, and other styles
Consult the top 50 journal articles for your research on the topic 'Power predictions.'
Next to every source in the list of references, there is an 'Add to bibliography' button. Press on it, and we will generate automatically the bibliographic reference to the chosen work in the citation style you need: APA, MLA, Harvard, Chicago, Vancouver, etc.
You can also download the full text of the academic publication as pdf and read online its abstract whenever available in the metadata.
Browse journal articles on a wide variety of disciplines and organise your bibliography correctly.
1
Beardsell, Alec. "Power predictions." Physics World 33, no. 5 (2020): 26. http://dx.doi.org/10.1088/2058-7058/33/5/25.
Full text
2
Zhuang, Wei, Zhiheng Li, Ying Wang, Qingyu Xi, and Min Xia. "GCN–Informer: A Novel Framework for Mid-Term Photovoltaic Power Forecasting." Applied Sciences 14, no. 5 (2024): 2181. http://dx.doi.org/10.3390/app14052181.
Full textAbstract:
Predicting photovoltaic (PV) power generation is a crucial task in the field of clean energy. Achieving high-accuracy PV power prediction requires addressing two challenges in current deep learning methods: (1) In photovoltaic power generation prediction, traditional deep learning methods often generate predictions for long sequences one by one, significantly impacting the efficiency of model predictions. As the scale of photovoltaic power stations expands and the demand for predictions increases, this sequential prediction approach may lead to slow prediction speeds, making it difficult to meet real-time prediction requirements. (2) Feature extraction is a crucial step in photovoltaic power generation prediction. However, traditional feature extraction methods often focus solely on surface features, and fail to capture the inherent relationships between various influencing factors in photovoltaic power generation data, such as light intensity, temperature, and more. To overcome these limitations, this paper proposes a mid-term PV power prediction model that combines Graph Convolutional Network (GCN) and Informer models. This fusion model leverages the multi-output capability of the Informer model to ensure the timely generation of predictions for long sequences. Additionally, it harnesses the feature extraction ability of the GCN model from nodes, utilizing graph convolutional modules to extract feature information from the ‘query’ and ‘key’ components within the attention mechanism. This approach provides more reliable feature information for mid-term PV power prediction, thereby ensuring the accuracy of long sequence predictions. Results demonstrate that the GCN–Informer model significantly reduces prediction errors while improving the precision of power generation forecasting compared to the original Informer model. Overall, this research enhances the prediction accuracy of PV power generation and contributes to advancing the field of clean energy.
3
Wu, Xinhua, Nan Chen, Qianyun Du, Shuangshuang Mao, and Xiaoming Ju. "Short-term wind power prediction model based on ARMA-GRU-QPSO and error correction." Journal of Physics: Conference Series 2427, no. 1 (2023): 012028. http://dx.doi.org/10.1088/1742-6596/2427/1/012028.
Full textAbstract:
Abstract Power system dispatch benefits from accurate wind power predictions. To increase the prediction precision for wind power, this paper proposes a combined model for predicting short-term wind power based on the autoregressive moving average-gated recurrent unit (ARMA-GRU). Firstly, we build the ARMA model and GRU model respectively to predict wind power. Then we optimize the combined model’s weights by quantum particle swarm algorithm (QPSO). Finally, we build an error correction model for the prediction errors to acquire the final results for the wind power predictions. Our experimental results prove the model’s reliability and the model’s high predictability is verified by comparing different prediction models.
4
Shen, Runjie, Ruimin Xing, Yiying Wang, Danqiong Hua, and Ming Ma. "Ultra-short-term prediction method of photovoltaic electric field power based on ground-based cloud image segmentation." E3S Web of Conferences 185 (2020): 01052. http://dx.doi.org/10.1051/e3sconf/202018501052.
Full textAbstract:
As a large number of photovoltaic power stations are built and put into operation, the total amount of photovoltaic power generation accounts for an increasing proportion of the total electricity. The inability to accurately predict solar energy output has brought great uncertainty to the grid. Therefore, predicting the future power of photovoltaic fields is of great significance. According to different time scales, predictions are divided into long-term, medium-term and ultra-short-term predictions. The main difficulty of ultra-short-term forecasting lies in the power fluctuations caused by sudden and drastic changes in environmental factors. The shading of clouds is directly related to the irradiance received on the surface of the photovoltaic panel, which has become the main factor affecting the fluctuation of photovoltaic power generation. Therefore, sky images captured by conventional cameras installed near solar panels can be used to analyze cloud characteristics and improve the accuracy of ultra-short-term predictions. This paper uses historical power information of photovoltaic power plants and cloud image data, combined with machine learning methods, to provide ultra-short-term predictions of the power generation of photovoltaic power plants. First, the random forest method is used to use historical power generation data to establish a single time series prediction model to predict ultra-short-term power generation. Compared with the continuous model, the root mean square (RMSE) error of prediction is reduced by 28.38%. Secondly, the Unet network is used to segment the cloud image, and the cloud amount information is analyzed and input into the random forest prediction model to obtain the bivariate prediction model. The experimental results prove that, based on the cloud amount information contained in the cloud chart, the bivariate prediction model has an 11.56% increase in prediction accuracy compared with the single time series prediction model, and an increase of 36.66% compared with the continuous model.
5
Jin, Xue-Bo, Hong-Xing Wang, Xiao-Yi Wang, Yu-Ting Bai, Ting-Li Su, and Jian-Lei Kong. "Deep-Learning Prediction Model with Serial Two-Level Decomposition Based on Bayesian Optimization." Complexity 2020 (September 14, 2020): 1–14. http://dx.doi.org/10.1155/2020/4346803.
Full textAbstract:
The power load prediction is significant in a sustainable power system, which is the key to the energy system’s economic operation. An accurate prediction of the power load can provide a reliable decision for power system planning. However, it is challenging to predict the power load with a single model, especially for multistep prediction, because the time series load data have multiple periods. This paper presents a deep hybrid model with a serial two‐level decomposition structure. First, the power load data are decomposed into components; then, the gated recurrent unit (GRU) network, with the Bayesian optimization parameters, is used as the subpredictor for each component. Last, the predictions of different components are fused to achieve the final predictions. The power load data of American Electric Power (AEP) were used to verify the proposed predictor. The results showed that the proposed prediction method could effectively improve the accuracy of power load prediction.
6
Maitanova, Nailya, Jan-Simon Telle, Benedikt Hanke, et al. "A Machine Learning Approach to Low-Cost Photovoltaic Power Prediction Based on Publicly Available Weather Reports." Energies 13, no. 3 (2020): 735. http://dx.doi.org/10.3390/en13030735.
Full textAbstract:
A fully automated transferable predictive approach was developed to predict photovoltaic (PV) power output for a forecasting horizon of 24 h. The prediction of PV power output was made with the help of a long short-term memory machine learning algorithm. The main challenge of the approach was using (1) publicly available weather reports without solar irradiance values and (2) measured PV power without any technical information about the PV system. Using this input data, the developed model can predict the power output of the investigated PV systems with adequate accuracy. The lowest seasonal mean absolute scaled error of the prediction was reached by maximum size of the training set. Transferability of the developed approach was proven by making predictions of the PV power for warm and cold periods and for two different PV systems located in Oldenburg and Munich, Germany. The PV power prediction made with publicly available weather data was compared to the predictions made with fee-based solar irradiance data. The usage of the solar irradiance data led to more accurate predictions even with a much smaller training set. Although the model with publicly available weather data needed greater training sets, it could still make adequate predictions.
7
Liu, Shipeng, Dejun Ning, and Jue Ma. "TCNformer Model for Photovoltaic Power Prediction." Applied Sciences 13, no. 4 (2023): 2593. http://dx.doi.org/10.3390/app13042593.
Full textAbstract:
Despite the growing capabilities of the short-term prediction of photovoltaic power, we still face two challenges to longer time-range predictions: error accumulation and long-term time series feature extraction. In order to improve the longer time range prediction accuracy of photovoltaic power, this paper proposes a seq2seq prediction model TCNformer, which outperforms other state-of-the-art (SOTA) algorithms by introducing variable selection (VS), long- and short-term time series feature extraction (LSTFE), and one-step temporal convolutional network (TCN) decoding. A VS module employs correlation analysis and periodic analysis to separate the time series correlation information, LSTFE extracts multiple time series features from time series data, and one-step TCN decoding realizes generative predictions. We demonstrate here that TCNformer has the lowest mean squared error (MSE), mean absolute error (MAE) and mean absolute percentage error (MAPE) in contrast to the other algorithms in the field of the short-term prediction of photovoltaic power, and furthermore, the effectiveness of each module has been verified through ablation experiments.
8
Guo, Wei, Li Xu, Tian Wang, Danyang Zhao, and Xujing Tang. "Photovoltaic Power Prediction Based on Hybrid Deep Learning Networks and Meteorological Data." Sensors 24, no. 5 (2024): 1593. http://dx.doi.org/10.3390/s24051593.
Full textAbstract:
Conventional point prediction methods encounter challenges in accurately capturing the inherent uncertainty associated with photovoltaic power due to its stochastic and volatile nature. To address this challenge, we developed a robust prediction model called QRKDDN (quantile regression and kernel density estimation deep learning network) by leveraging historical meteorological data in conjunction with photovoltaic power data. Our aim is to enhance the accuracy of deterministic predictions, interval predictions, and probabilistic predictions by incorporating quantile regression (QR) and kernel density estimation (KDE) techniques. The proposed method utilizes the Pearson correlation coefficient for selecting relevant meteorological factors, employs a Gaussian Mixture Model (GMM) for clustering similar days, and constructs a deep learning prediction model based on a convolutional neural network (CNN) combined with a bidirectional gated recurrent unit (BiGRU) and attention mechanism. The experimental results obtained using the dataset from the Australian DKASC Research Centre unequivocally demonstrate the exceptional performance of QRKDDN in deterministic, interval, and probabilistic predictions for photovoltaic (PV) power generation. The effectiveness of QRKDDN was further validated through ablation experiments and comparisons with classical machine learning models.
9
Cahyadi, Catra Indra, Suwarno Suwarno, Aminah Asmara Dewi, Musri Kona, Muhammad Arif, and Muhammad Caesar Akbar. "Solar Prediction Strategy for Managing Virtual Power Stations." International Journal of Energy Economics and Policy 13, no. 4 (2023): 503–12. http://dx.doi.org/10.32479/ijeep.14124.
Full textAbstract:
The potential for solar power is available in Indonesia because it is located on the equator, with good sunshine all year round. The Indonesian government is currently actively developing a solar power plant while still looking at the consequences of development, especially on the surrounding environment. It is necessary to pay attention so that it does not disturb the surrounding environment, which can also cause climate change. The city of Medan is one of the largest cities in Indonesia, which has direct exposure to sunlight which is quite promising for predicting solar power plants in the future. Solar energy generation in the last decade has continued to improve and develop in solar power predictions in a short period. Integration of solar power sources without accurate power predictions can hinder network operations and the use of renewable generation sources. To solve this problem, virtual power plant modeling can solve as a solution that manages to minimize the prediction error. This research studies methods that can efficiently generate significant daily Photovoltaic (PV) predictions at the locations studied using available data from the Meteorology, Climatology, and Geophysics Agency (MCGA). The approach of the two models based on RMSE (root mean square error) and MAE (Mean Absolute Error), can be virtual power plant modeling can solve it as a management solution to be minimal in its prediction error. This research studies methods that can efficiently generate significant daily Photovoltaic (PV) predictions at the locations studied using available data from the Meteorology, Climatology, and Geophysics Agency (MCGA). The approach of the two models based on RMSE (root mean square error) and MAE (Mean Absolute Error), can be virtual power plant modeling can solve it as a management solution to be minimal in its prediction error. This research studies methods that can efficiently generate significant daily Photovoltaic (PV) predictions at the locations studied using available data from the Meteorology, Climatology, and Geophysics Agency (MCGA). The approach of the two models based on RMSE (root mean square error) and MAE (Mean Absolute Error), can be taking into account the uncertainty of predictions that can provide additional information from virtual power plants. Verified prediction strategy performance against PV module power output and a set based on geographic meteorological station data have been used to simulate Virtual Power Plants (VPP). The power forecasting prediction refers to the LSTM (Long Short-Term Memory) network and gives an error close to that of other learning methods, based on the RMS characteristic of 4.19 W/m2 under lead time with different launch times. The application of the VPP model can reduce the global error by about 12.37% with the model RMSE, and shows great potential.
10
Xhaferaj, Blenard. "INVESTIGATION ON SOME CONVENTIONAL HULLS FORMS OF THE PREDICTIVE ACCURACY OF A PARAMETRIC SOFTWARE FOR PRELIMINARY PREDICTIONS OF RESISTANCE AND POWER." Brodogradnja 73, no. 1 (2022): 1–22. http://dx.doi.org/10.21278/brod73101.
Full textAbstract:
Preliminary prediction of resistance and power is a fundamental aspect of the ship design process since they directly influence the developments of the design process, fuel consumption and costs, and environmental impact from the early design stage. Parametric predictions of resistance and power, based mainly on statistical regression models that are also ideal for computer programming, are often performed during initial design stages, providing rapid predictions and optimisations for minimum resistance. The paper aims to present the results of the comparative analysis on some conventional hulls of the predictive accuracy of a computer program developed by the author for parametric predictions of resistance and power of ships. The program (entitled Ship Power V 1.0) is developed in the Visual Basic 6.0 environment based on two well-known regression models Holtrop and Van Oortmerssen. The program can perform detailed predictions of resistance and power, resistance coefficients, propeller thrust, hull efficiency, wake, and trust fractions, with no restriction on the number of velocities. In this study, only the analysis of the accuracy of resistance and power prediction is considered. Results of the comparative analysis of the computational procedures of Ship Power V 1.0 versus experimental data, and against results of another well-known commercial software, performed on three models of the Ridgely-Nevitt trawler series and KCS hull have shown a good level of accuracy and reliability as other well-known commercial software.
11
Liu, Yi, Jun He, Yu Wang, Zong Liu, Lixun He, and Yanyang Wang. "Short-Term Wind Power Prediction Based on CEEMDAN-SE and Bidirectional LSTM Neural Network with Markov Chain." Energies 16, no. 14 (2023): 5476. http://dx.doi.org/10.3390/en16145476.
Full textAbstract:
Accurate wind power data prediction is crucial to increase wind energy usage since wind power data are characterized by uncertainty and randomness, which present significant obstacles to the scheduling of power grids. This paper proposes a hybrid model for wind power prediction based on complementary ensemble empirical mode decomposition with adaptive noise (CEEMDAN), sample entropy (SE), bidirectional long short-term memory network (BiLSTM), and Markov chain (MC). First, CEEMDAN is used to decompose the wind power series into a series of subsequences at various frequencies, and then SE is employed to reconstruct the wind power series subsequences to reduce the model’s complexity. Second, the long short-term memory (LSTM) network is optimized, the BiLSTM neural network prediction method is used to predict each reconstruction component, and the results of the different component predictions are superimposed to acquire the total prediction results. Finally, MC is used to correct the model’s total prediction results to increase the accuracy of the predictions. Experimental validation with measured data from wind farms in a region of Xinjiang, and computational results demonstrate that the proposed model can better fit wind power data than other prediction models and has greater prediction accuracy and generalizability for enhancing wind power prediction performance.
12
Lange, Matthias. "On the Uncertainty of Wind Power Predictions—Analysis of the Forecast Accuracy and Statistical Distribution of Errors." Journal of Solar Energy Engineering 127, no. 2 (2005): 177–84. http://dx.doi.org/10.1115/1.1862266.
Full textAbstract:
In this work the uncertainty of wind power predictions is investigated with a special focus on the important role of the nonlinear power curve. Based on numerical predictions and measured data from six onshore wind farms the overall prediction accuracy is assessed and it is shown that due to the power curve the relative forecast error increases by a factor of 1.8–2.6 compared to the wind speed forecast. This factor can be considered as an effective nonlinearity factor. A decomposition of the commonly known root mean square error is beneficially used to distinguish different error sources related to either on-site conditions or global properties of the numerical weather prediction system. The statistical distribution of the wind speed prediction error is found to be Gaussian in contrast to the the one of power prediction error. Using the power curve and conditional probability density functions of the wind speed the unsymmetric distribution of the power prediction error can be explained and modeled such that it can be estimated even if no measurement data is available.
13
Horat, Nina, Sina Klerings, and Sebastian Lerch. "Improving Model Chain Approaches for Probabilistic Solar Energy Forecasting through Post-processing and Machine Learning." Advances in Atmospheric Sciences 42, no. 2 (2024): 297–312. https://doi.org/10.1007/s00376-024-4219-2.
Full textAbstract:
AbstractWeather 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.
14
Hu, Hongda, Zhiyong Hu, Kaiwen Zhong, et al. "Long-term offshore wind power prediction using spatiotemporal kriging: A case study in China’s Guangdong Province." Energy Exploration & Exploitation 38, no. 3 (2019): 703–22. http://dx.doi.org/10.1177/0144598719889368.
Full textAbstract:
The predicted wind power in coastal waters is an important factor when planning and developing offshore wind farms. The stochastic wind field challenges the accuracy of these predictions. Using single-point wind measurements, most previous studies have focused on the prediction of short-term wind power, ranging from minutes to several days. Longer-term wind power predictions would better support decision-making related to offshore wind power balance management and reserve capacities. In addition, larger-scale wind power predictions, based on gridded wind field data, would provide a more comprehensive understanding of the spatiotemporal variations of wind energy resources. In this study, a spatiotemporal ordinary kriging model was developed to predict the offshore wind power density on a monthly basis using the cross-calibrated multiplatform gridded wind field data. The spatiotemporal variations of wind power density were directly quantified through the development of spatiotemporal variograms that integrated spatial and temporal distances. The proposed model achieved a notable performance with an overall R2 of 0.94 and a relative prediction error of 16.35% in the validation experiment of predicting the monthly wind power density from 2013 in the coastal waters of China’s Guangdong Province. Using this model, the spatial distributions of wind power density along Guangdong’s coastal waters at monthly, seasonal, and annual time-scales from 2013 were accurately predicted. The experiment results demonstrated the remarkable potential of the spatiotemporal ordinary kriging model to provide reliable long-term prediction for offshore wind energy resources.
15
Song, Yujeong, Jisu Park, Myoung-Seok Suh, and Chansoo Kim. "Prediction of Full-Load Electrical Power Output of Combined Cycle Power Plant Using a Super Learner Ensemble." Applied Sciences 14, no. 24 (2024): 11638. https://doi.org/10.3390/app142411638.
Full textAbstract:
Combined Cycle Power Plants (CCPPs) generate electrical power through gas turbines and use the exhaust heat from those turbines to power steam turbines, resulting in 50% more power output compared to traditional simple cycle power plants. Predicting the full-load electrical power output () of a CCPP is crucial for efficient operation and sustainable development. Previous studies have used machine learning models, such as the Bagging and Boosting models to predict . In this study, we propose employing Super Learner (SL), an ensemble machine learning algorithm, to enhance the accuracy and robustness of predictions. SL utilizes cross-validation to estimate the performance of diverse machine learning models and generates an optimal weighted average based on their respective predictions. It may provide information on the relative contributions of each base learner to the overall prediction skill. For constructing the SL, we consider six individual and ensemble machine learning models as base learners and assess their performances compared to the SL. The dataset used in this study was collected over six years from an operational CCPP. It contains one output variable and four input variables: ambient temperature, atmospheric pressure, relative humidity, and vacuum. The results show that the Boosting algorithms significantly influence the performance of the SL in comparison to the other base learners. The SL outperforms the six individual and ensemble machine learning models used as base learners. It indicates that the SL improves the generalization performance of predictions by combining the predictions of various machine learning models.
16
Yulianto, Tri Wahyu, Unit Three Kartini, and Bambang Suprianto. "Design of Forecasting Electrical Power of Ultra-Short-Term Solar Power Using the Hybrid Model K-Nearest Neighbors LSTM." Jurnal Indonesia Sosial Teknologi 5, no. 7 (2024): 3412–22. http://dx.doi.org/10.59141/jist.v5i7.1230.
Full textAbstract:
For the application of renewable energy at the airport, the use of solar power requires certainty of the electricity produced. The certainty of electricity generated from solar power can be predicted using machine learning methods. Predictions made on PV electrical power output are based on historical data from direct measurements from solar PV parameters, including solar radiation and PV panel temperature. Various types of machine learning methods for predicting PV output power have been used in previous studies with different eval_uation values of prediction results. In this study, the author conducted a hybrid K-NN method with LSTM to predict the PV electrical power of solar PV output with solar radiation parameters and PV panel temperature. After making predictions using this method, excellent RSME results were obtained with a value of 0.015424830635781967. The results of the PV output power value graph in this prediction are also very good, where the predicted value is close to the value of the testing data or actual data.
17
Liu, Renfeng, Yinbo Song, Chen Yuan, Desheng Wang, Peihua Xu, and Yaqin Li. "GAN-Based Abrupt Weather Data Augmentation for Wind Turbine Power Day-Ahead Predictions." Energies 16, no. 21 (2023): 7250. http://dx.doi.org/10.3390/en16217250.
Full textAbstract:
This study introduces a data augmentation technique based on generative adversarial networks (GANs) to improve the accuracy of day-ahead wind power predictions. To address the peculiarities of abrupt weather data, we propose a novel method for detecting mutation rates (MR) and local mutation rates (LMR). By analyzing historical data, we curated datasets that met specific mutation rate criteria. These transformed wind speed datasets were used as training instances, and using GAN-based methodologies, we generated a series of augmented training sets. The enriched dataset was then used to train the wind power prediction model, and the resulting prediction results were meticulously evaluated. Our empirical findings clearly demonstrate a significant improvement in the accuracy of day-ahead wind power prediction due to the proposed data augmentation approach. A comparative analysis with traditional methods showed an approximate 5% increase in monthly average prediction accuracy. This highlights the potential of leveraging mutated wind speed data and GAN-based techniques for data augmentation, leading to improved accuracy and reliability in wind power predictions. In conclusion, this paper presents a robust data augmentation method for wind power prediction, contributing to the potential enhancement of day-ahead prediction accuracy. Future research could explore additional mutation rate detection methods and strategies to further enhance GAN models, thereby amplifying the effectiveness of wind power prediction.
18
Korkmaz, Kadir Burak, Sofia Werner, and Rickard Bensow. "Verification and Validation of CFD Based Form Factors as a Combined CFD/EFD Method." Journal of Marine Science and Engineering 9, no. 1 (2021): 75. http://dx.doi.org/10.3390/jmse9010075.
Full textAbstract:
Predicting the propulsive power of ships with high accuracy still remains a challenge. Well established practices in the 1978 ITTC Power Prediction method have been questioned such as the form factor approach and its determination method. This paper investigates the possibility to improve the power predictions by the introduction of a combined CFD/EFD Method where the experimental determination of form factor is replaced by double body RANS computations. Following the Quality Assurance Procedure proposed by ITTC, a best practice guideline has been derived for the CFD based form factor determination method by applying systematic variations to the CFD set-ups. Following the verification and validation of the CFD based form factor method in model scale, the full scale speed-power-rpm relations between large number of speed trials and full scale predictions using the CFD based form factors in combination with ITTC-57 line and numerical friction lines are investigated. It is observed that the usage of CFD based form factors improves the predictions in general and no deterioration is noted within the limits of this study. Therefore, the combination of EFD and CFD is expected to provide immediate improvements to the 1978 ITTC Performance Prediction Method.
19
Lee, Dongkyu, Jinhwa Jeong, Sung Hoon Yoon, and Young Tae Chae. "Improvement of Short-Term BIPV Power Predictions Using Feature Engineering and a Recurrent Neural Network." Energies 12, no. 17 (2019): 3247. http://dx.doi.org/10.3390/en12173247.
Full textAbstract:
The time resolution and prediction accuracy of the power generated by building-integrated photovoltaics are important for managing electricity demand and formulating a strategy to trade power with the grid. This study presents a novel approach to improve short-term hourly photovoltaic power output predictions using feature engineering and machine learning. Feature selection measured the importance score of input features by using a model-based variable importance. It verified that the normative sky index in the weather forecasted data had the least importance as a predictor for hourly prediction of photovoltaic power output. Six different machine-learning algorithms were assessed to select an appropriate model for the hourly power output prediction with onsite weather forecast data. The recurrent neural network outperformed five other models, including artificial neural networks, support vector machines, classification and regression trees, chi-square automatic interaction detection, and random forests, in terms of its ability to predict photovoltaic power output at an hourly and daily resolution for 64 tested days. Feature engineering was then used to apply dropout observation to the normative sky index from the training and prediction process, which improved the hourly prediction performance. In particular, the prediction accuracy for overcast days improved by 20% compared to the original weather dataset used without dropout observation. The results show that feature engineering effectively improves the short-term predictions of photovoltaic power output in buildings with a simple weather forecasting service.
20
Huang, Zhaowen. "Research on Artificial Intelligence Methods for Feature Extraction and Prediction of Current Fluctuations." Applied and Computational Engineering 116, no. 1 (2024): 143–47. https://doi.org/10.54254/2755-2721/2025.20424.
Full textAbstract:
Accurate prediction of current fluctuations in power systems is crucial for ensuring the safety and stability of grid operations. Traditional prediction methods face significant limitations when dealing with data complexity and dynamic changes. The introduction of artificial intelligence technologies offers a new perspective for current fluctuation predictions. Based on Long Short-Term Memory networks (LSTM), this study proposes an intelligent prediction method combining deep fusion and feature extraction of multi-source data. This method processes multi-dimensional information such as historical current data, weather conditions, and load demands to achieve high-accuracy and real-time predictions of current fluctuations. The results show that this method significantly improves prediction accuracy, enhances model adaptability, and provides reliable support for power dispatch and resource optimization, offering important reference value for the development of smart grids.
21
Chui, Juanita Noeline, and Keith Ong. "Improving the prediction of effective lens position for intraocular lens power calculations." Asian Journal of Ophthalmology 17, no. 2 (2020): 233–42. http://dx.doi.org/10.35119/asjoo.v17i2.585.
Full textAbstract:
Purpose: Achieving the desired post-operative refraction in cataract surgery requires accurate calculations for intraocular lens (IOL) power. Latest-generation formulae use anterior-chamber depth (ACD)—the distance from the corneal apex to the anterior surface of the lens—as one of the parameters to predict the post-operative IOL position within the eye, termed the effective lens position (ELP). Significant discrepancies between predicted and actual ELP result in refractive surprise. This study aims to improve the predictability of ELP. We hypothesise that predictions based on the distance from the corneal apex to the mid-sagittal plane of the cataractous lens would more accurately reflect the position of the principal plane of the non-angulated IOL within the capsular bag. Accordingly, we propose that predictions derived from ACD + ½LT (length thickness) would be superior to those from ACD alone.
 Design: Retrospective cohort study, comparing ELP predictions derived from ACD to aproposed prediction parameter.
 Method: This retrospective study includes data from 162 consecutive cataract surgery cases, with posterior-chamber IOL (AlconSN60WF) implantation. Pre- and postoperative biometric measurements were made using the IOLMaster700 (ZEISS, Jena, Germany). The accuracy and reliability of ELP predictions derived from ACD and ACD + ½LT were compared using software-aided analyses.
 Results: An overall reduction in average ELP prediction error (PEELP) was achieved using the proposed parameter (root-mean-square-error [RMSE] = 0.50 mm), compared to ACD (RMSE = 1.57 mm). The mean percentage PEELP, comparing between eyes of different axial lengths, was 9.88% ± 3.48% and −34.9% ± 4.79% for predictions derived from ACD + ½LT and ACD, respectively. A 44.10% ± 5.22% mean of differences was observed (p < 0.001).
 Conclusion: ACD + ½LT predicts ELP with greater accuracy and reliability than ACD alone; its use in IOL power calculation formulae may improve refractive outcomes.
22
Brown, Nathan P., Tommy Ao, Daniel H. Dolan, Marcus D. Knudson, and J. Matthew D. Lane. "DENNIS: a design and analysis tool for dynamic material x-ray diffraction experiments." Journal of Instrumentation 19, no. 07 (2024): P07030. http://dx.doi.org/10.1088/1748-0221/19/07/p07030.
Full textAbstract:
Abstract We present DENNIS (Diffraction Experiment desigN and aNalysiS): a graphical software tool useful for the design and analysis of dynamic x-ray diffraction experiments, such as those performed on the Z Pulsed Power Facility, Thor Pulsed Power Generator, and Dynamic Compression Sector (DCS) of the Advanced Photon Source. DENNIS provides rapid powder and single-crystal diffraction pattern predictions and powder diffraction pattern image integration in three-dimensional geometries. Additional features include crystallographic information file reading, image processing, and synthetic diffraction pattern image generation. We overview the software's capabilities, detail the prediction and integration methodologies, and provide example implementations on Z and DCS experiments.
23
XU, Yiquan, Jinyu GUO, Rui LUO, et al. "Refinement method of equivalent source power level determination for traffic noise in urban noise mapping." INTER-NOISE and NOISE-CON Congress and Conference Proceedings 270, no. 10 (2024): 1925–36. http://dx.doi.org/10.3397/in_2024_3105.
Full textAbstract:
For traffic noise prediction, the predictive models of source power level determination commonly in use are the Technical Guidelines for Noise Impact Assessment in China or the Calculation of Road Traffic Noise model in the United Kingdom, which incorporate traffic flow and car speeds. However, the accuracy of these models is often compromised due to urban traffic's complexity, vehicle diversity, and unforeseen events. This study investigates a refinement method based on noise on-site measurement and aforementioned approaches for equivalent source power level determination for traffic noise in urban noise mapping. In the study, onsite measurements are conducted to collect noise and traffic data, and source power level calculated by refinement method is validated using measurement data. Different prediction methods are employed for measurement sites calculations. By comparisons between predictions with noise on-site measurement data, the refinement method proposed in the study demonstrates a significant accuracy enhancement in traffic noise prediction over previous models for determining equivalent source power level. This enhancement highlights the efficacy of adopting noise measurement-based methods for setting traffic noise source power levels, leading to more precise traffic noise predictions in complex urban settings, particularly in areas surrounding urban arterial roads.
24
Guo, Xianchao, Yuchang Mo, and Ke Yan. "Short-Term Photovoltaic Power Forecasting Based on Historical Information and Deep Learning Methods." Sensors 22, no. 24 (2022): 9630. http://dx.doi.org/10.3390/s22249630.
Full textAbstract:
The accurate prediction of photovoltaic (PV) power is essential for planning power systems and constructing intelligent grids. However, this has become difficult due to the intermittency and instability of PV power data. This paper introduces a deep learning framework based on 7.5 min-ahead and 15 min-ahead approaches to predict short-term PV power. Specifically, we propose a hybrid model based on singular spectrum analysis (SSA) and bidirectional long short-term memory (BiLSTM) networks with the Bayesian optimization (BO) algorithm. To begin, the SSA decomposes the PV power series into several sub-signals. Then, the BO algorithm automatically adjusts hyperparameters for the deep neural network architecture. Following that, parallel BiLSTM networks predict the value of each component. Finally, the prediction of the sub-signals is summed to generate the final prediction results. The performance of the proposed model is investigated using two datasets collected from real-world rooftop stations in eastern China. The 7.5 min-ahead predictions generated by the proposed model can reduce up to 380.51% error, and the 15 min-ahead predictions decrease by up to 296.01% error. The experimental results demonstrate the superiority of the proposed model in comparison to other forecasting methods.
25
Kmen, Christopher, Gerhard Navratil, and Ioannis Giannopoulos. "Location, Location, Location: The Power of Neighborhoods for Apartment Price Predictions Based on Transaction Data." ISPRS International Journal of Geo-Information 13, no. 12 (2024): 425. http://dx.doi.org/10.3390/ijgi13120425.
Full textAbstract:
Land and real estate have long been regarded as stable investments, with property prices steadily rising, underscoring the need for accurate predictive models to capture the varying rates of price growth across different locations. This study leverages a decade-long dataset of 83,527 apartment transactions in Vienna, Austria, to train machine learning models using XGBoost. Unlike most prior research, the extended time span of the dataset enables predictions for multiple future years, providing a more robust long-term prediction. The primary objective is to examine how spatial factors can enhance real estate price predictions. In addition to transaction data, socio-demographic and geographic variables were collected to characterize the neighborhoods surrounding each apartment. Ten models, each varying in the number of input years, were trained to predict the price per square meter. The model performance was assessed using the mean absolute percentage error (MAPE), offering insights into their predictive accuracy for both short-term and long-term predictions. This study underscores the importance of distinguishing between newly built and existing apartments in real estate price modeling. By splitting the dataset prior to training, predictive models focusing solely on newly built properties achieved an average reduction of about 6% in MAPE. The best-performing models achieved an average MAPE of 15% for one-year-ahead predictions and maintained a MAPE below 20% for predictions up to three years ahead, demonstrating the effectiveness of leveraging spatial features to enhance real estate price prediction accuracy.
26
Liu, Hai Ke, Jiang Xia Feng, Shen Quan Yang, and Tao Jia. "Wind Power Prediction Model Based on ARMA and Improved BP-ANN." Advanced Materials Research 1008-1009 (August 2014): 183–87. http://dx.doi.org/10.4028/www.scientific.net/amr.1008-1009.183.
Full textAbstract:
In order to improve the prediction accuracy of wind power, this research is based on time series and improved BP-ANN algorithm. The basic idea is described as follows: wind speed forecasting model is established by using time series method; wind speed-wind power model is built by utilising improved BP-ANN algorithm; wind speed data from time series forecasting is used as input of neural network model, and the prediction results for wind power are obtained. In order to analyse the availability of wind power prediction model, the mean absolute error and correlation coefficient are compared to analyse the predictions results. The results show that the prediction model can effectively improve the forecasting accuracy of wind power.
27
Peng, Yan, Shichen Wang, Wenjin Chen, Junchao Ma, Chenxu Wang, and Jingwei Chen. "LightGBM-Integrated PV Power Prediction Based on Multi-Resolution Similarity." Processes 11, no. 4 (2023): 1141. http://dx.doi.org/10.3390/pr11041141.
Full textAbstract:
Improving the accuracy of PV power prediction is conducive to PV participation in economic dispatch and power market transactions in the distribution network, as well as safe dispatch and operation of the grid. Considering that the selection of highly correlated historical data can improve the accuracy of PV power prediction, this study proposes an integrated PV power prediction method based on a multi-resolution similarity consideration that considers both trend similarity and detail similarity. Firstly, using irradiance as the similarity variable, similar-days were selected using grey correlation analysis to form a set of similar data to control the similarity, with the overall trend of the day to be predicted at a macro level. Using irradiance to calculate the similarity at each specific point in time via Euclidean distance, similar-times were identified to form another set of similar data to consider the degree of similarity in detail. The above approach enables the selection of similarity data for both resolutions. Then, a 1DCNN-LSTM prediction model that considers the feature correlation of different variables and the temporal dependence of a single variable was proposed. Three important features were selected by a random forest model as inputs to the prediction model, and two similar data training models with different resolutions were used to generate a photovoltaic power prediction model based on similar-days and similar-times. Ultimately, the learning of the two predictions integrated with LightGBM compensate for each other, generating highly accurate predictions that combine the advantages of multi-resolution similarity considerations. Actual operation data of a PV power station was used for verification. The simulation results show that the prediction effect of ensemble learning was better than that of the single 1DCNN-LSTM model. The proposed method was compared with other commonly used PV power prediction models. In the data case of this study, it was found that the proposed method reduced the prediction error rate by 1.48%, 11.4%, and 6.45%, compared to the LSTM, CNN, and BP, respectively. Experiments show that model prediction results considering the selection of similar data at multiple resolutions can provide more extensive information to an ensemble learner and reduce the deviation in model predictions. Therefore, the proposed method can provide a reference for PV integration into the grid and participation in market-based electricity trading, which is of great significance.
28
Gao, Jinming, Xianlong Su, Changsu Kim, Kerang Cao, and Hoekyung Jung. "A Parallel Prediction Model for Photovoltaic Power Using Multi-Level Attention and Similar Day Clustering." Energies 17, no. 16 (2024): 3958. http://dx.doi.org/10.3390/en17163958.
Full textAbstract:
Photovoltaic (PV) power generation is significantly impacted by environmental factors that exhibit substantial uncertainty and volatility, posing a critical challenge for accurate PV power prediction in power system management. To address this, a parallel model is proposed for PV short-term prediction utilizing a multi-level attention mechanism. Firstly, gray relation analysis (GRA) and an improved ISODATA algorithm are used to select a dataset of similar days with comparable meteorological characteristics to the forecast day. A transformer encoder layer with multi-head attention is then used to extract long-term dependency features. Concurrently, BiGRU, optimized with a Global Attention network, is used to capture global temporal features. Feature fusion is performed using Cross Attention, calculating attention weights to emphasize significant features and enhancing feature integration. Finally, high-precision predictions are achieved through a fully connected layer. Utilizing historical PV power generation data to predict power output under various weather conditions, the proposed model demonstrates superior performance across all three climate types compared to other models, achieving more reliable predictions.
29
Al-Dahidi, Sameer, Osama Ayadi, Jehad Adeeb, Mohammad Alrbai, and Bashar R. Qawasmeh. "Extreme Learning Machines for Solar Photovoltaic Power Predictions." Energies 11, no. 10 (2018): 2725. http://dx.doi.org/10.3390/en11102725.
Full textAbstract:
The unpredictability of intermittent renewable energy (RE) sources (solar and wind) constitutes reliability challenges for utilities whose goal is to match electricity supply to consumer demands across centralized grid networks. Thus, balancing the variable and increasing power inputs from plants with intermittent energy sources becomes a fundamental issue for transmission system operators. As a result, forecasting techniques have obtained paramount importance. This work aims at exploiting the simplicity, fast computational and good generalization capability of Extreme Learning Machines (ELMs) in providing accurate 24 h-ahead solar photovoltaic (PV) power production predictions. The ELM architecture is firstly optimized, e.g., in terms of number of hidden neurons, and number of historical solar radiations and ambient temperatures (embedding dimension) required for training the ELM model, then it is used online to predict the solar PV power productions. The investigated ELM model is applied to a real case study of 264 kWp solar PV system installed on the roof of the Faculty of Engineering at the Applied Science Private University (ASU), Amman, Jordan. Results showed the capability of the ELM model in providing predictions that are slightly more accurate with negligible computational efforts compared to a Back Propagation Artificial Neural Network (BP-ANN) model, which is currently adopted by the PV system owners for the prediction task.
30
Putintseva, Maria. "PREDICTIVE POWER OF INFORMATION MARKET PRICES." Journal of Prediction Markets 5, no. 2 (2012): 44–74. http://dx.doi.org/10.5750/jpm.v5i2.489.
Full textAbstract:
Prediction (or information) markets are markets where participants trade contracts whose payoff depends on unknown future events. Studying prediction markets allows to avoid many problems, which arise in some artificially designed behavioral experiments investigating collective decision making or individual's belief formation. This work is aimed, first, to verify whether predictions made by prices of binary options traded in information markets are reliable and whether the prices contain additional information about the future comparing to the information available from the dynamics of underlying asset only. Second, inter- and intraday microstructure of the market of binary options on Dow Jones Industrial Average index is examined and described quantitatively. Third, since some ability to forecast future changes in the underlying asset is detected, a simple trading strategy based on observing the trading process in the prediction market is suggested and its profitability and applicability is evaluated.
31
Li, Fan, Hongzhen Wang, Dan Wang, Dong Liu, and Ke Sun. "A Review of Wind Power Prediction Methods Based on Multi-Time Scales." Energies 18, no. 7 (2025): 1713. https://doi.org/10.3390/en18071713.
Full textAbstract:
In response to the ‘zero carbon’ goal, the development of renewable energy has become a global consensus. Among the array of renewable energy sources, wind energy is distinguished by its considerable installed capacity on a global scale. Accurate wind power prediction provides a fundamental basis for power grid dispatching, unit combination operation, and wind farm operation and maintenance. This study establishes a framework to bridge theoretical innovations with practical implementation challenges in wind power prediction. This work uses a narrative method to synthesize and discuss wind power prediction methods. Common classification angles of wind power prediction methods are outlined. By synthesizing existing approaches through multi-time scales, from the ultra-short term and short term to mid-long term, the review further deconstructs methods by model characteristics, input data types, spatial scales, and evaluation metrics. The analysis reveals that the data-driven prediction model dominates ultra-short-term predictions through rapid response to volatility, while the hybrid method enhances short-term precision. Mid-term predictions increasingly integrate climate dynamics to address seasonal variability. A key contribution lies in unifying fragmented methodologies into a decision support framework that prioritizes the time scale, model adaptability, and spatial constraints. This work enables practitioners to systematically select optimal strategies and advance the development of forecasting systems that are critical for highly renewable energy systems.
32
Sun, Ziwen, Tao Wang, Yanning Lu, et al. "A Novel Condenser Vacuum Degree Prediction Model Based on LSTM and MemN2N." Journal of Physics: Conference Series 2294, no. 1 (2022): 012030. http://dx.doi.org/10.1088/1742-6596/2294/1/012030.
Full textAbstract:
Abstract Condenser vacuum degree prediction of power plants is a challenge task in power system security field. Most existing studies are based on shallow machine learning algorithms, which fail to leverage historical data comprehensively, resulting inaccuracy and unreliable predictions. Therefore, using a serialization model like Recurrent Neural Network to capture time-series information from historical data is necessary. However, these serialization model alone has inherent defects in dealing with long-distance dependence, which may cause historical information forgetting problem. This paper proposes a new prediction model combining LSTM and End-To-End Memory Network (MemN2N). We use LSTM to mine the long-distance dependency information in historical data, and introduce the encoding historical information into the memory pool of MemN2N. MemN2N allows better preservation of historical information for serialization model LSTM, and can make accurate and reliable predictions through soft attention mechanism. Through the experiments on real data from the power plant show that, compared with other prediction models, the model proposed in this paper achieves higher prediction accuracy and has great engineering value.
33
Ciechulski, Tomasz. "Forecasting of short-term power demands in Polish Power System using ensemble of LSTM networks." Journal of Automation, Electronics and Electrical Engineering 7, no. 1 (2025): 29–37. https://doi.org/10.24136/jaeee.2025.003.
Full textAbstract:
The article presents and discusses the results of the research of forecasting power demands in Polish Power System with time horizon of one hour ahead in conditions of limited availability of forecasting model input data, covering only three months. The prediction was carried out using deep neural networks - LSTM (Long Short-Term Memory) connected to an ensemble. The performance of the ensemble is much more efficient than individual networks working separately. The numerical experiments were conducted using MATLAB computing environment. The accuracy of the predictions was estimated using such statistical measures as MAPE, MAE, RMSE, Pearson correlation coefficient R.
34
Gao, Li, Hong, and Long. "Short-Term Forecasting of Power Production in a Large-Scale Photovoltaic Plant Based on LSTM." Applied Sciences 9, no. 15 (2019): 3192. http://dx.doi.org/10.3390/app9153192.
Full textAbstract:
Photovoltaic (PV) power is attracting more and more concerns. Power output prediction, as a necessary technical requirement of PV plants, closely relates to the rationality of power grid dispatch. If the accuracy of power prediction in PV plants can be further enhanced by forecasting, stability of power grid will be improved. Therefore, a 1-h-ahead power output forecasting based on long-short-term memory (LSTM) networks is proposed. The forecasting output of the model is based on the time series of 1-h-ahead numerical weather prediction to reveal the spatio-temporal characteristic. The comprehensive meteorological conditions, including different types of season and weather conditions, were considered in the model, and parameters of LSTM models were investigated simultaneously. Analysis of prediction result reveals that the proposed model leads to a superior prediction performance compared with traditional PV output power predictions. The accuracy of output power prediction is enhanced by 3.46–13.46%.
35
Frid, S. E., N. V. Lisitskaya, and O. S. Popel. "Results of applicability analysis of satellite observations and reanalysis data for autonomous photovoltaic unit simulation." Доклады Академии наук 488, no. 6 (2019): 609–11. http://dx.doi.org/10.31857/s0869-56524886609-611.
Full textAbstract:
The accuracy analysis of energy performance prediction of autonomous photovoltaic units using various climate databases (NASA POWER, SARAH-E, CLARA-A, ERA5, Meteonorm, etc.) for some geographic points in Russia was performed by comparing with calculations using data of World Radiation Data Center. It is shown that the considered databases provide a spread of predictions of the required power of solar battery at the level of 10-20% only when solar fraction is less than 70%. For larger solar fraction, the prediction error of the required power of solar battery can reach hundreds of percent.
36
Villarini, Gabriele, and Gabriel A. Vecchi. "Multiseason Lead Forecast of the North Atlantic Power Dissipation Index (PDI) and Accumulated Cyclone Energy (ACE)." Journal of Climate 26, no. 11 (2013): 3631–43. http://dx.doi.org/10.1175/jcli-d-12-00448.1.
Full textAbstract:
Abstract By considering the intensity, duration, and frequency of tropical cyclones, the power dissipation index (PDI) and accumulated cyclone energy (ACE) are concise metrics routinely used to assess tropical storm activity. This study focuses on the development of a hybrid statistical–dynamical seasonal forecasting system for the North Atlantic Ocean’s PDI and ACE over the period 1982–2011. The statistical model uses only tropical Atlantic and tropical mean sea surface temperatures (SSTs) to describe the variability exhibited by the observational record, reflecting the role of both local and nonlocal effects on the genesis and development of tropical cyclones in the North Atlantic basin. SSTs are predicted using a 10-member ensemble of the Geophysical Fluid Dynamics Laboratory Climate Model, version 2.1 (GFDL CM2.1), an experimental dynamical seasonal-to-interannual prediction system. To assess prediction skill, a set of retrospective predictions is initialized for each month from November to April, over the years 1981–2011. The skill assessment indicates that it is possible to make skillful predictions of ACE and PDI starting from November of the previous year: skillful predictions of the seasonally integrated North Atlantic tropical cyclone activity for the coming season could be made even while the current one is still under way. Probabilistic predictions for the 2012 North Atlantic tropical cyclone season are presented.
37
MARTINEZ GARRIZ, IÑAKI, PILAR HERRERA PLAZA, and MAIALEN LARRETXEA URRUTIA. "N-BIR: A NUMERIC OPTIMIZATION APPROACH FOR POWER ELECTRONIC CONVERTER BURN-IN TESTING TIME REDUCTION." DYNA 99, no. 2 (2024): 201–7. http://dx.doi.org/10.6036/10866.
Full textAbstract:
Burn-in testing is an effective method for detecting early faults in electronic products before they reach the market. This test has a high cost due to lengthy test time on a test bench. In this paper, we propose N-BIR (Numeric optimization approach for power electronic Burn-In testing time Reduction), an algorithm capable of predicting the burn-in (BI) test temperature of power electronic converters, intending to shorten the duration of such tests. This algorithm optimizes by least squares a theoretical model of the system, using as data a fraction of the total burn-in test. Moreover, not only is it capable of making accurate predictions, but it also accompanies them with a prediction interval (PI), so that the algorithm itself can quantify how confident it is of its predictions. We show that using 40% of a conventional rolling test total, our algorithm outperforms several of today's most common Machine Learning (ML) algorithms. Furthermore, we show that it can reduce burning time by 50% to 60% by making accurate predictions, which makes it possible to identify a significant portion of converters that don't require full testing, ultimately lowering costs and boosting productivity. Keywords: Burn-in temperature prediction, Burn-in time reduction, Levelized Cost of Energy, Machine Learning, Electronic Power Converter, Reliability.
38
Baran, Sándor, and Ágnes Baran. "Calibration of wind speed ensemble forecasts for power generation." Időjárás 125, no. 4 (2021): 609–24. http://dx.doi.org/10.28974/idojaras.2021.4.4.
Full textAbstract:
In the last decades, wind power became the second largest energy source in the EU covering 16% of its electricity demand. However, due to its volatility, accurate short range wind power predictions are required for successful integration of wind energy into the electrical grid. Accurate predictions of wind power require accurate hub height wind speed forecasts, where the state-of-the-art method is the probabilistic approach based on ensemble forecasts obtained from multiple runs of numerical weather prediction models. Nonetheless, ensemble forecasts are often uncalibrated and might also be biased, thus require some form of post-processing to improve their predictive performance. We propose a novel flexible machine learning approach for calibrating wind speed ensemble forecasts, which results in a truncated normal predictive distribution. In a case study based on 100m wind speed forecasts produced by the operational ensemble prediction system of the Hungarian Meteorological Service, the forecast skill of this method is compared with the predictive performance of three different ensemble model output statistics approaches and the raw ensemble forecasts. We show that compared with the raw ensemble, post-processing always improves the calibration of probabilistic and accuracy of point forecasts, and from the four competing methods, the novel machine learning based approach results in the best overall performance.
39
Diqi, Mohammad, Ahmad Wakhid, I. Wayan Ordiyasa, Nurhadi Wijaya, and Marselina Endah Hiswati. "Harnessing the Power of Stacked GRU for Accurate Weather Predictions." Indonesian Journal of Artificial Intelligence and Data Mining 6, no. 2 (2023): 208. http://dx.doi.org/10.24014/ijaidm.v6i2.24769.
Full textAbstract:
This research proposed a novel approach using Stacked GRU (Gated Recurrent Unit) models to address the problem of weather prediction and aimed to improve forecasting accuracy in sectors like agriculture, transportation, and disaster management. The key idea involved leveraging the temporal dependencies and memory management capabilities of Stacked GRU to model complex weather patterns effectively. Comprehensive data preprocessing ensured data quality and fine-tuning of the model architecture and hyperparameters optimized performance. The research demonstrated the Stacked GRU model's effectiveness in accurately forecasting temperature, pressure, humidity, and wind speed, validated by low RMSE and MAE scores and high R2 coefficients. However, challenges in forecasting humidity and a percentage discrepancy in wind speed predictions were observed. Overfitting and computational complexity were identified as potential limitations. Despite these constraints, the study concluded that the Stacked GRU model showed promise in weather forecasting and warranted further refinement for broader applications in time-series prediction tasks.
40
Taleb, Ihab, Guillaume Guerard, Frédéric Fauberteau, and Nga Nguyen. "A Flexible Deep Learning Method for Energy Forecasting." Energies 15, no. 11 (2022): 3926. http://dx.doi.org/10.3390/en15113926.
Full textAbstract:
Load prediction with higher accuracy and less computing power has become an important problem in the smart grids domain in general and especially in demand-side management (DSM), as it can serve to minimize global warming and better integrate renewable energies. To this end, it is interesting to have a general prediction model which uses different standard machine learning models in order to be flexible enough to be used in different regions and/or countries and to give a prediction for multiple days or weeks with relatively good accuracy. Thus, we propose in this article a flexible hybrid machine learning model that can be used to make predictions of different ranges by using both standard neural networks and an automatic process of updating the weights of these models depending on their past errors. The model was tested on Mayotte Island and the mean absolute percentage error (MAPE) obtained was 1.71% for 30 min predictions, 3.5% for 24 h predictions, and 5.1% for one-week predictions.
41
Lee, B. E., C. A. J. Fletcher, and M. Behnia. "Computational Prediction of Tube Erosion in Coal Fired Power Utility Boilers." Journal of Engineering for Gas Turbines and Power 121, no. 4 (1999): 746–50. http://dx.doi.org/10.1115/1.2818536.
Full textAbstract:
Erosion of boiler tubes causes serious operational problems in many pulverized coal-fired utility boilers. A new erosion model has been developed in the present study for the prediction of boiler tube erosion. The Lagrangian approach is employed to predict the behavior of the particulate phase. The results of computational prediction of boiler tube erosion and the various parameters causing erosion are discussed in this paper. Comparison of the numerical predictions for a single tube erosion with experimental data shows very good agreement.
42
Ramadevi, Bhukya, Venkata Ramana Kasi, and Kishore Bingi. "Hybrid LSTM-Based Fractional-Order Neural Network for Jeju Island’s Wind Farm Power Forecasting." Fractal and Fractional 8, no. 3 (2024): 149. http://dx.doi.org/10.3390/fractalfract8030149.
Full textAbstract:
Efficient integration of wind energy requires accurate wind power forecasting. This prediction is critical in optimising grid operation, energy trading, and effectively harnessing renewable resources. However, the wind’s complex and variable nature poses considerable challenges to achieving accurate forecasts. In this context, the accuracy of wind parameter forecasts, including wind speed and direction, is essential to enhancing the precision of wind power predictions. The presence of missing data in these parameters further complicates the forecasting process. These missing values could result from sensor malfunctions, communication issues, or other technical constraints. Addressing this issue is essential to ensuring the reliability of wind power predictions and the stability of the power grid. This paper proposes a long short-term memory (LSTM) model to forecast missing wind speed and direction data to tackle these issues. A fractional-order neural network (FONN) with a fractional arctan activation function is also developed to enhance generated wind power prediction. The predictive efficacy of the FONN model is demonstrated through two comprehensive case studies. In the first case, wind direction and forecast wind speed data are used, while in the second case, wind speed and forecast wind direction data are used for predicting power. The proposed hybrid neural network model improves wind power forecasting accuracy and addresses data gaps. The model’s performance is measured using mean errors and R2 values.
43
Metaxas, Phillip, and Andrew Leigh. "The Predictive Power of Political Pundits: Prescient or Pitiful?" Media International Australia 147, no. 1 (2013): 5–17. http://dx.doi.org/10.1177/1329878x1314700103.
Full textAbstract:
Although Australian political pundits frequently make predictions about the future, little systematic evidence exists about the accuracy of these predictions. To assess the predictive power of experts, we survey the transcripts of two well-known political programs – Insiders and Meet the Press – and record all falsifiable forecasts. Looking at the three months prior to both the 2007 and 2010 federal elections, we are struck by the paucity of falsifiable predictions, with most pundits heavily qualifying their predictions (so that they can never be said to be wrong). In 32 hours of television, we identify 20 falsifiable forecasts in our sample, of which we judge thirteen to be correct. We conclude with some suggestions for political talk shows, and for political scientists seeking to better analyse expert predictions.
44
Baillie, Emma, Piers D. L. Howe, Andrew Perfors, Tim Miller, Yoshihisa Kashima, and Andreas Beger. "Explainable models for forecasting the emergence of political instability." PLOS ONE 16, no. 7 (2021): e0254350. http://dx.doi.org/10.1371/journal.pone.0254350.
Full textAbstract:
Building on previous research on the use of macroeconomic factors for conflict prediction and using data on political instability provided by the Political Instability Task Force, this article proposes two minimal forecasting models of political instability optimised to have the greatest possible predictive power for one-year and two-year event horizons, while still making predictions that are fully explainable. Both models employ logistic regression and use just three predictors: polity code (a measure of government type), infant mortality, and years of stability (i.e., years since the last instability event). These models make predictions for 176 countries on a country-year basis and achieve AUPRC’s of 0.108 and 0.115 for the one-year and two-year models respectively. They use public data with ongoing availability so are readily reproducible. They use Monte Carlo simulations to construct confidence intervals for their predictions and are validated by testing their predictions for a set of reference years separate from the set of reference years used to train them. This validation shows that the models are not overfitted but suggests that some of the previous models in the literature may have been. The models developed in this article are able to explain their predictions by showing, for a given prediction, which predictors were the most influential and by using counterfactuals to show how the predictions would have been altered had these predictors taken different values. These models are compared to models created by lasso regression and it is shown that they have at least as good predictive power but that their predictions can be more readily explained. Because policy makers are more likely to be influenced by models whose predictions can explained, the more interpretable a model is the more likely it is to influence policy.
45
MARUŠIĆ, M., and S. VUK-PAVLOVIĆ. "PREDICTION POWER OF MATHEMATICAL MODELS FOR TUMOR GROWTH." Journal of Biological Systems 01, no. 01 (1993): 69–78. http://dx.doi.org/10.1142/s0218339093000069.
Full textAbstract:
We compared the Gompertz model, the generalized Gompertz model, the Piantadosi model, the autostimulation model and the polynomials for the power to predict growth of multicellular tumor spheroids as paradigms of the prevascular phase of tumor growth. For the comparison of models we developed a criterion that established the Gompertz model as the model with the best prediction power. The prediction power of the remaining models was ranked in declining order: the generalized Gompertz model; the mutually indistinguishable Piantadosi model and the autostimulation model; and the polynomials. The ranking of models was not affected by the applied minimization criteria of weighted least squares, unweighted least squares and fitting to logarithmically transformed data, but the prediction power was affected by these criteria. The best predictions were obtained by weighted least squares, closely followed by fitting to logarithmically transformed data. The unweighted least-squares minimization was much less applicable for prediction (and description) of growth.
46
Farrell, Alayna, Jennifer King, Caroline Draxl, et al. "Design and analysis of a wake model for spatially heterogeneous flow." Wind Energy Science 6, no. 3 (2021): 737–58. http://dx.doi.org/10.5194/wes-6-737-2021.
Full textAbstract:
Abstract. Methods of turbine wake modeling are being developed to more accurately account for spatially variant atmospheric conditions within wind farms. Most current wake modeling utilities are designed to apply a uniform flow field to the entire domain of a wind farm. When this method is used, the accuracy of power prediction and wind farm controls can be compromised depending on the flow-field characteristics of a particular area. In an effort to improve strategies of wind farm wake modeling and power prediction, FLOw Redirection and Induction in Steady State (FLORIS) was developed to implement sophisticated methods of atmospheric characterization and power output calculation. In this paper, we describe an adapted FLORIS model that features spatial heterogeneity in flow-field characterization. This model approximates an observed flow field by interpolating from a set of atmospheric measurements that represent local weather conditions. The objective of this method is to capture heterogeneous atmospheric effects caused by site-specific terrain features, without explicitly modeling the geometry of the wind farm terrain. The implemented adaptations were validated by comparing the simulated power predictions generated from FLORIS to the actual recorded wind farm output from the supervisory control and data acquisition (SCADA) recordings and large eddy simulations (LESs). When comparing the performance of the proposed heterogeneous model to homogeneous FLORIS simulations, the results show a 14.6 % decrease for mean absolute error (MAE) in wind farm power output predictions for cases using wind farm SCADA data and a 18.9 % decrease in LES case studies. The results of these studies also indicate that the efficacy of the proposed modeling techniques may vary with differing site-specific operational conditions. This work quantifies the accuracy of wind plant power predictions under heterogeneous flow conditions and establishes best practices for atmospheric surveying for wake modeling.
47
Desell, Travis J., AbdElRahman A. ElSaid, Zimeng Lyu, David Stadem, Shuchita Patwardhan, and Steve Benson. "Long term predictions of coal fired power plant data using evolved recurrent neural networks." at - Automatisierungstechnik 68, no. 2 (2020): 130–39. http://dx.doi.org/10.1515/auto-2019-0116.
Full textAbstract:
AbstractThis work presents an investigation into the ability of recurrent neural networks (RNNs) to provide long term predictions of time series data generated by coal fired power plants. While there are numerous studies which have used artificial neural networks (ANNs) to predict coal plant parameters, to the authors’ knowledge these have almost entirely been restricted to predicting values at the next time step, and not farther into the future. Using a novel neuro-evolution strategy called Evolutionary eXploration of Augmenting Memory Models (EXAMM), we evolved RNNs with advanced memory cells to predict per-minute plant parameters and per-hour boiler parameters up to 8 hours into the future. These data sets were challenging prediction tasks as they involve spiking behavior in the parameters being predicted. While the evolved RNNs were able to successfully predict the spikes in the hourly data they did not perform very well in accurately predicting their severity. The per-minute data proved even more challenging as medium range predictions miscalculated the beginning and ending of spikes, and longer range predictions reverted to long term trends and ignored the spikes entirely. We hope this initial study will motivate further study into this highly challenging prediction problem. The use of fuel properties data generated by a new Coal Tracker Optimization (CTO) program was also investigated and this work shows that their use improved predictive ability of the evolved RNNs.
48
Garcia, J., F. J. Casson, L. Frassinetti, et al. "Modelling performed for predictions of fusion power in JET DTE2: overview and lessons learnt." Nuclear Fusion 63, no. 11 (2023): 112003. http://dx.doi.org/10.1088/1741-4326/acedc0.
Full textAbstract:
Abstract For more than a decade, an unprecedented predict-first activity has been carried in order to predict the fusion power and provide guidance to the second Deuterium–Tritium (D–T) campaign performed at JET in 2021 (DTE2). Such an activity has provided a framework for a broad model validation and development towards the D–T operation. It is shown that it is necessary to go beyond projections using scaling laws in order to obtain detailed physics based predictions. Furthermore, mixing different modelling complexity and promoting an extended interplay between modelling and experiment are essential towards reliable predictions of D–T plasmas. The fusion power obtained in this predict-first activity is in broad agreement with the one finally measured in DTE2. Implications for the prediction of fusion power in future devices, such as ITER, are discussed.
49
GUPTA, BHASKAR SEN, and SHANKAR P. DAS. "TESTING POWER-LAW RELAXATION SCENARIOS IN A METASTABLE LIQUID." International Journal of Modern Physics B 26, no. 29 (2012): 1250146. http://dx.doi.org/10.1142/s0217979212501469.
Full textAbstract:
The renormalized dynamics described by the equations of nonlinear fluctuating hydrodynamics (NFH) treated at one loop order gives rise to the basic model of the mode coupling theory (MCT). We investigate here by analyzing the density correlation function, a crucial prediction of ideal MCT, namely the validity of the multi step relaxation scenario. The equilibrium density correlation function is calculated here from the direct solutions of NFH equations for a hard sphere system. We make first detailed investigation for the robustness of the correlation functions obtained from the numerical solutions by varying the size of the grid. For an optimum choice of grid size we analyze the decay of the density correlation function to identify the multi-step relaxation process. Weak signatures of two step power law relaxation is seen with exponents which do not match predictions from the one loop MCT. For the final relaxation stretched exponential (KWW) behavior is seen and the relaxation time grows with increase of density. But apparent power law divergences indicate a critical packing fraction much higher than the corresponding MCT predictions for a hard sphere fluid.
50
Pallarés, Jesus G., Jose R. Lillo-Bevia, Ricardo Morán-Navarro, Victor Cerezuela-Espejo, and Ricardo Mora-Rodriguez. "Time to exhaustion during cycling is not well predicted by critical power calculations." Applied Physiology, Nutrition, and Metabolism 45, no. 7 (2020): 753–60. http://dx.doi.org/10.1139/apnm-2019-0637.
Full textAbstract:
Three to 5 cycling tests to exhaustion allow prediction of time to exhaustion (TTE) at power output based on calculation of critical power (CP). We aimed to determine the accuracy of CP predictions of TTE at power outputs habitually endured by cyclists. Fourteen endurance-trained male cyclists underwent 4 randomized cycle-ergometer TTE tests at power outputs eliciting (i) mean Wingate anaerobic test (WAnTmean), (ii) maximal oxygen consumption, (iii) respiratory compensation threshold (VT2), and (iv) maximal lactate steady state (MLSS). Tests were conducted in duplicate with coefficient of variation of 5%–9%. Power outputs were 710 ± 63 W for WAnTmean, 366 ± 26 W for maximal oxygen consumption, 302 ± 31 W for VT2 and 247 ± 20 W for MLSS. Corresponding TTE were 00:29 ± 00:06, 03:23 ± 00:45, 11:29 ± 05:07, and 76:05 ± 13:53 min:s, respectively. Power output associated with CP was only 2% lower than MLSS (242 ± 19 vs. 247 ± 20 W; P < 0.001). The CP predictions overestimated TTE at WAnTmean (00:24 ± 00:10 mm:ss) and MLSS (04:41 ± 11:47 min:s), underestimated TTE at VT2 (–04:18 ± 03:20 mm:ss; P < 0.05), and correctly predicted TTE at maximal oxygen consumption. In summary, CP accurately predicts MLSS power output and TTE at maximal oxygen consumption. However, it should not be used to estimate time to exhaustion in trained cyclists at higher or lower power outputs (e.g., sprints and 40-km time trials). Novelty CP calculation enables to predict TTE at any cycling power output. We tested those predictions against measured TTE in a wide range of cycling power outputs. CP appropriately predicted TTE at maximal oxygen consumption intensity but err at higher and lower cycling power outputs.