scholarly journals A Probabilistic Ensemble Prediction Method for PV Power in the Nonstationary Period

Energies ◽  
2021 ◽  
Vol 14 (4) ◽  
pp. 859
Author(s):  
Yuan An ◽  
Kaikai Dang ◽  
Xiaoyu Shi ◽  
Rong Jia ◽  
Kai Zhang ◽  
...  
Keyword(s):  
Prediction Accuracy ◽  
Ensemble Prediction ◽  
Virtual Power ◽  
Short Term ◽  
Virtual Power Plant ◽  
Interval Prediction ◽  
Photovoltaic Power ◽  
Proposed Model ◽  
Prediction Module ◽  
Stationary Period

Due to the large number of grid connection of distributed power supply, the existing scheduling methods can not meet the demand gradually. The proposed virtual power plant provides a new idea to solve this problem. The photovoltaic power prediction provides the data basis for the scheduling of the virtual power plant. Prediction intervals of photovoltaic power is a powerful statistical tool used for quantifying the uncertainty of photovoltaic power generation in power systems. To improve the interval prediction accuracy during the non-stationary periods of photovoltaic power, this paper proposes a probabilistic ensemble prediction model, which combines the modules of data preprocessing, non-stationary period discrimination, feature extraction, deterministic prediction, uncertainty prediction, and optimization integration into a general framework. More specifically, in the non-stationary period discrimination module, the method of discriminating the difference of the power ratio difference is introduced and applied for identifying the non-stationary period of the data of photovoltaic output; in the deterministic point prediction module, a stacking- long-short-term memory neural network model is used for point forecasts; in the uncertainty interval prediction module, a BAYES neural network is introduced for probabilistic forecasts; in the optimization integration module, an optimization algorithm named Non-dominated Sorting Genetic Algorithm-II is applied for integrating and optimizing the results of the point forecast and probabilistic forecast. The proposed model is tested using two photovoltaic outputs and weather data measured from a grid-connected photovoltaic system. The results show that the proposed model outperforms conventional forecast methods to predict short-term photovoltaic power outputs and associated uncertainties. The interval width is reduced by 10–20%, and the prediction accuracy is improved by at least 10%; this can be a useful tool for photovoltaic power forecasting.

A novel fuzzy rough set based long short-term memory integration model for energy consumption prediction of public buildings

2020 ◽  
pp. 1-15
Author(s):  
Hongchang Sun ◽  
Yadong wang ◽  
Lanqiang Niu ◽  
Fengyu Zhou ◽  
Heng Li
Keyword(s):  
Energy Consumption ◽  
Rough Set ◽  
Prediction Accuracy ◽  
Short Term Memory ◽  
Short Term ◽  
Fuzzy Rough Set ◽  
Proposed Model ◽  
Long Short Term Memory ◽  
Energy Consumption Prediction ◽  
Consumption Prediction

Building energy consumption (BEC) prediction is very important for energy management and conservation. This paper presents a short-term energy consumption prediction method that integrates the Fuzzy Rough Set (FRS) theory and the Long Short-Term Memory (LSTM) model, and is thus named FRS-LSTM. This method can find the most directly related factors from the complex and diverse factors influencing the energy consumption, which improves the prediction accuracy and efficiency. First, the FRS is used to reduce the redundancy of the input features by the attribute reduction of the factors affecting the energy consumption forecasting, and solves the data loss problem caused by the data discretization of a classical rough set. Then, the final attribute set after reduction is taken as the input of the LSTM networks to obtain the final prediction results. To validate the effectiveness of the proposed model, this study used the actual data of a public building to predict the building’s energy consumption, and compared the proposed model with the LSTM, Levenberg-Marquardt Back Propagation (LM-BP), and Support Vector Regression (SVR) models. The experimental results reveal that the presented FRS-LSTM model achieves higher prediction accuracy compared with other comparative models.

scholarly journals Forecasting Day-Ahead Hourly Photovoltaic Power Generation Using Convolutional Self-Attention Based Long Short-Term Memory

Energies ◽  
2020 ◽  
Vol 13 (15) ◽  
pp. 4017 ◽  
Author(s):  
Dukhwan Yu ◽  
Wonik Choi ◽  
Myoungsoo Kim ◽  
Ling Liu
Keyword(s):  
Power Generation ◽  
Deep Learning ◽  
Prediction Accuracy ◽  
Short Term Memory ◽  
Local Context ◽  
Short Term ◽  
Term Memory ◽  
Proposed Model ◽  
Long Short Term Memory ◽  
Pv Power Generation

The problem of Photovoltaic (PV) power generation forecasting is becoming crucial as the penetration level of Distributed Energy Resources (DERs) increases in microgrids and Virtual Power Plants (VPPs). In order to improve the stability of power systems, a fair amount of research has been proposed for increasing prediction performance in practical environments through statistical, machine learning, deep learning, and hybrid approaches. Despite these efforts, the problem of forecasting PV power generation remains to be challenging in power system operations since existing methods show limited accuracy and thus are not sufficiently practical enough to be widely deployed. Many existing methods using long historical data suffer from the long-term dependency problem and are not able to produce high prediction accuracy due to their failure to fully utilize all features of long sequence inputs. To address this problem, we propose a deep learning-based PV power generation forecasting model called Convolutional Self-Attention based Long Short-Term Memory (LSTM). By using the convolutional self-attention mechanism, we can significantly improve prediction accuracy by capturing the local context of the data and generating keys and queries that fit the local context. To validate the applicability of the proposed model, we conduct extensive experiments on both PV power generation forecasting using a real world dataset and power consumption forecasting. The experimental results of power generation forecasting using the real world datasets show that the MAPEs of the proposed model are much lower, in fact by 7.7%, 6%, 3.9% compared to the Deep Neural Network (DNN), LSTM and LSTM with the canonical self-attention, respectively. As for power consumption forecasting, the proposed model exhibits 32%, 17% and 44% lower Mean Absolute Percentage Error (MAPE) than the DNN, LSTM and LSTM with the canonical self-attention, respectively.

scholarly journals A load-balancing power scheduling system for virtual power plant considering emission reduction and charging demand of moving electric vehicles

2019 ◽  
Vol 52 (5-6) ◽  
pp. 687-701
Author(s):  
Chenn-Jung Huang ◽  
An-Feng Liu ◽  
Kai-Wen Hu ◽  
Liang-Chun Chen ◽  
Yu-Kang Huang
Keyword(s):  
Power Generation ◽  
Power Plant ◽  
Energy Management ◽  
Energy Sources ◽  
Virtual Power ◽  
Short Term ◽  
Scheduling System ◽  
Virtual Power Plant ◽  
Power Scheduling ◽  
Term Energy

With the rapid development of the emerging technologies and significant cost reduction of the deployment for solar energy and wind power, the replacement of traditional power generation by renewable energy becomes feasible in the future. However, different from currently deployed centralized power sources, renewables are categorized as one kind of intermittent energy sources, and the scale of renewables is small and scattered. In the recent literature, the architecture of virtual power plant was proposed to replace the current smart grid in the future. However, the energy sharing concept and the uncertainties of intermittent energy sources will cause the short-term energy management for the virtual power plant much more complicated than the current centralized control energy management for traditional power generation system. We thus propose a hierarchical day-ahead power scheduling system for virtual power plant in this work to tackle the complex short-term energy management problems. We first collect electricity consumption data from smart appliances used in households and predict power-generating capacity of renewable energy sources at the prosumer level. Then, the proposed hierarchical power scheduling system is employed to schedule the usage of electricity for the customers by considering the efficiency of the use of distributed renewables. Notably, charging management of a moving electric vehicle is also considered in the proposed power scheduling mechanism. In addition, a real-time power tracking mechanism is presented to deal with the forecast errors of volatile renewable power generation, electricity load, and moving electric vehicle charging, and the maximal usage of renewables and reduction of the burden on community virtual power plants during time period of peak load can be achieved accordingly. The experimental results show that the proposed day-ahead power scheduling system can mitigate the dependency on traditional power generation effectively, and balance peak and off-peak period load of electricity market.

scholarly journals Improvement of Short-Term BIPV Power Predictions Using Feature Engineering and a Recurrent Neural Network

Energies ◽  
2019 ◽  
Vol 12 (17) ◽  
pp. 3247 ◽  
Author(s):  
Dongkyu Lee ◽  
Jinhwa Jeong ◽  
Sung Hoon Yoon ◽  
Young Tae Chae
Keyword(s):  
Neural Network ◽  
Machine Learning ◽  
Recurrent Neural Network ◽  
Power Output ◽  
Prediction Accuracy ◽  
Support Vector ◽  
Feature Engineering ◽  
Short Term ◽  
Interaction Detection ◽  
Photovoltaic Power

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.

scholarly journals Attention-Based Long Short-Term Memory Recurrent Neural Network for Capacity Degradation of Lithium-Ion Batteries

Batteries ◽  
2021 ◽  
Vol 7 (4) ◽  
pp. 66
Author(s):  
Tadele Mamo ◽  
Fu-Kwun Wang
Keyword(s):  
Neural Network ◽  
Recurrent Neural Network ◽  
Prediction Accuracy ◽  
Short Term Memory ◽  
Lithium Ion ◽  
Short Term ◽  
Term Memory ◽  
Proposed Model ◽  
Capacity Degradation ◽  
Long Short Term Memory

Monitoring cycle life can provide a prediction of the remaining battery life. To improve the prediction accuracy of lithium-ion battery capacity degradation, we propose a hybrid long short-term memory recurrent neural network model with an attention mechanism. The hyper-parameters of the proposed model are also optimized by a differential evolution algorithm. Using public battery datasets, the proposed model is compared to some published models, and it gives better prediction performance in terms of mean absolute percentage error and root mean square error. In addition, the proposed model can achieve higher prediction accuracy of battery end of life.

Optimal Scheduling of a Virtual Power Plant with Demand Response in Short-Term Electricity Market

2020 ◽  
Author(s):  
Homa Rashidizadeh-Kermani ◽  
Mostafa Vahedipour-Dahraie ◽  
Miadreza Shafie-khah ◽  
Gerardo J. Osorio ◽  
Joao P. S. Catalao
Keyword(s):  
Power Plant ◽  
Demand Response ◽  
Electricity Market ◽  
Optimal Scheduling ◽  
Virtual Power ◽  
Short Term ◽  
Virtual Power Plant

scholarly journals Short-Term Photovoltaic Power Generation Forecasting Based on Multivariable Grey Theory Model with Parameter Optimization

2017 ◽  
Vol 2017 ◽  
pp. 1-9 ◽  
Author(s):  
Zhifeng Zhong ◽  
Chenxi Yang ◽  
Wenyang Cao ◽  
Chenyang Yan
Keyword(s):  
Power Generation ◽  
Particle Swarm Optimization ◽  
Prediction Accuracy ◽  
Particle Swarm Optimization Algorithm ◽  
Theory Model ◽  
Grey Model ◽  
Grey Theory ◽  
Short Term ◽  
Swarm Optimization ◽  
Photovoltaic Power

Owing to the environment, temperature, and so forth, photovoltaic power generation volume is always fluctuating and subsequently impacts power grid planning and operation seriously. Therefore, it is of great importance to make accurate prediction of the power generation of photovoltaic (PV) system in advance. In order to improve the prediction accuracy, in this paper, a novel particle swarm optimization algorithm based multivariable grey theory model is proposed for short-term photovoltaic power generation volume forecasting. It is highlighted that, by integrating particle swarm optimization algorithm, the prediction accuracy of grey theory model is expected to be highly improved. In addition, large amounts of real data from two separate power stations in China are being employed for model verification. The experimental results indicate that, compared with the conventional grey model, the mean relative error in the proposed model has been reduced from 7.14% to 3.53%. The real practice demonstrates that the proposed optimization model outperforms the conventional grey model from both theoretical and practical perspectives.

Sign in / Sign up

Export Citation Format

Share Document