Search

The biggest wind farm of Turkey is placed at Manisa which is located in the Aegean Region. Electricity is a non-storable commodity for that reason, it is very important to have a strong forecast and model of the potential electricity production to plan the electricity loads. In this study, the aim is to model and forecast electricity production of the wind farms located at Manisa by using a univariate model based on sequence-to-sequence learning. The forecasting range of the study is from short term to midterm. The strength of the proposed model is that; it only needs its own lagged value to make forecasts. The empirical evidences show that the model has high coefficient of variation (R2) in short term and moderate R2 in the midterm forecast. Although in the mid-range forecasts R2 slightly decreases mean squared error and mean absolute error shows that the model is accurate also in the midterm forecasts. The proposed model is not only strong in hourly electricity production forecasts but with a slight modification also in forecasting the minimum, maximum and average electricity production for a fixed range. This study concludes with two fresh and intriguing future research ideas. [Display omitted] [ABSTRACT FROM AUTHOR]

Wind power represents a promising source of renewable energies. Precise forecasting of wind power generation is crucial to mitigate the challenges of balancing supply and demand in the smart grid. Nevertheless, the major difficulty in wind power is its high fluctuation and intermittent nature, making it challenging to forecast. This study aims to develop efficient data-driven models to accurately forecast wind power generation. Crucially, the main contributions of this work are listed in the following major elements. Firstly, we investigate the performance of enhanced machine learning models to forecast univariate wind power time-series data. Specifically, we employed Bayesian optimization (BO) to optimally tune hyperparameters of the Gaussian process regression (GPR), Support Vector Regression (SVR) with different kernels, and ensemble learning (ES) models (i.e., Boosted trees and Bagged trees) and investigated their forecasting performance. Secondly, dynamic information has been incorporated in their construction to further enhance the forecasting performance of the investigated models. Specifically, we introduce lagged measurements to enable capturing time evolution into the design of the considered models. Furthermore, more input variables (e.g., wind speed and wind direction) are used to further improve wind prediction performance. Actual measurements from three wind turbines in France, Turkey, and Kaggle are used to verify the efficiency of the considered models. The results reveal the benefit of considering lagged data and input variables to better forecast wind power. The results also showed that the optimized GPR and ensemble models outperformed the other machine learning models. [ABSTRACT FROM AUTHOR]

Short-term electricity load forecasting plays a vital role in determining the energy supplied. It has daily, weekly, and yearly harmonics patterns. In Muslim countries, it is influenced by religious celebrations held on the Islamic Hijri calendar. This study aims to compare Dynamic Harmonic Regression forecasting and its hybrid version involving calendar variation effect. This study is designed in two scenarios. Scenario 1 shows that out-of-sample forecasting can capture both seasonal and calendar variation patterns, while Scenario 2 shows the performance while only seasonal patterns exist. The results show that the hybrid model outperforms the individual models in two scenarios. [ABSTRACT FROM AUTHOR]

Long-term load forecasting is a significant and complex topic in electric distribution systems. Forecasters is need to proper forecasting methodologies and smart solutions to minimize complexity. In this study, regional long-term load forecasting is presented, for Sivas province of Turkey, taking into account the development plan of the municipality, and subscriber profiles. Firstly, the municipality development plan is divided into regions of similar load characteristics. The load demand values of each region are defined mathematically using the S curve. The optimal parameter values of the S curve are calculated using meta-heuristic methods such as Genetic Algorithm (GA), Grey Wolf Optimization (GWO) and Harris Hawk Optimization (HHO). The obtained results are compared with the results of the econometrics-based (top-to-bottom) approach and actual consumer projection. The consumption values between 2004 and 2014 are used for parameter estimation of S curves. The consumption values obtained as the result of analysis the period between 2015 and 2018 were selected as test data. The result is shown that S curve-based regional demand forecast demonstrated more convenient results using the HHO algorithm with statistical values of RE = 1.3362, MAE = 1.5145, RMSE = 1.80385 and STD = 2.122 can be applied to the forecast regional electricity consumption. The proposed method is simple and can be easily applied to forecast the total consumption of the power load for a province any load forecasting region. The presented approach can be used to define the future projections of electricity distribution systems and determine the correct investment strategies. [ABSTRACT FROM AUTHOR]

In this study, a novel photovoltaic power forecasting system that utilizes a deep Convolutional Neural Network (CNN) structure and an input signal decomposition algorithm is proposed. The proposed CNN architecture extracts deep features to forecast short-term power using transfer learning-based AlexNet. The historical power, solar radiation, wind speed, and temperature data are selected as the input. The signal decomposition algorithm called Empirical Mode Decomposition (EMD) is utilized to decompose the historical power signal into sub-components. In order to extract deep features, all input parameters are converted to 2D feature maps and feed to the input of the CNN. The experiments are realized on a grid-tied Photovoltaic Power Plant (PVPP) that has 1000 kW installed capacity located in Turkey. The experiments are performed under four weather conditions as partial cloudy, cloudy-rainy, heavy-rainy, and sunny days to show the effectiveness of the proposed method. The obtained results are compared with the benchmark regression algorithms. When the results are analyzed, the proposed method gives the highest Correlation Coefficient (R) and the lowest Root Mean Square Error (RMSE), Mean Absolute Error (MAE), and SMAPE values under all horizons and weather conditions. For 1-h to 5-h ahead, the average R values of the proposed method are obtained as 97.28%, 95.77%, 94.49%, 93.61%, and 92.62%, respectively. The average RMSE values are observed as 4.90%, 6.30%, 7.50%, 8.00%, and 9.17% for 1-h to 5-h ahead. The experimental results confirm that the proposed method outperforms the conventional regression algorithms and reveals effective results with its competitive performance. [ABSTRACT FROM AUTHOR]

Short-term load forecasting models play a critical role in distribution companies in making effective decisions in their planning and scheduling for production and load balancing. Unlike aggregated load forecasting at the distribution level or substations, forecasting load profiles of many end-users at the customer-level, thanks to smart meters, is a complicated problem due to the high variability and uncertainty of load consumptions as well as customer privacy issues. In terms of customers' short-term load forecasting, these models include a high level of nonlinearity between input data and output predictions, demanding more robustness, higher prediction accuracy, and generalizability. In this paper, we develop an advanced preprocessing technique coupled with a hybrid sequential learning-based energy forecasting model that employs a convolution neural network (CNN) and bidirectional long short-term memory (BLSTM) within a unified framework for accurate energy consumption prediction. The energy consumption outliers and feature clustering are extracted at the advanced preprocessing stage. The novel hybrid deep learning approach based on data features coding and decoding is implemented in the prediction stage. The proposed approach is tested and validated using real-world datasets in Turkey, and the results outperformed the traditional prediction models compared in this paper. [ABSTRACT FROM AUTHOR]

Due to the coronavirus, millions of people worldwide carry out their work, education, shopping, culture, and entertainment activities from their homes now using the advantages of today's technology. Apart from this, patient care and follow-up are carried out with the help of electronic equipment especially in the institutions where health services are provided. It is important to provide a reliable electricity supply for humanity so that people can perform all these services. In this study, the outlook of energy in Turkey was examined. The current energy consumption and investments were examined. Then, the precautions by the government in the pandemic period according to the occurrence and spread of COVID-19 in the country are given in chronological order. The actual electricity consumption data were obtained daily across the country, after all these precautions. It was observed that electricity consumption decreased significantly, especially on restricted days. It is inarguable that energy consumption estimation should be made in the short term so that the energy sector is not adversely affected by this situation. In this study, more accurate short-term consumption forecasting methods were developed during the COVID-19 pandemic period: nonlinear autoregressive (NARX) and long short-term memory (LSTM) artificial neural networks (ANNs). Between January and April 2019 electrical consumption data were used to train and validate the forecast prediction. The NARX and LSTM are potential candidates for effective forecasting of electricity consumption. However, the obtained LSTM results suggest that the proposed method performs better than the NARX ANN. [ABSTRACT FROM AUTHOR]

In this study, wind power forecasting is performed for a Wind Power Plant (WPP) with an installed capacity of 135 MW in Turkey. The ruggedness index (RIX) of the terrain where WPP was installed is analyzed with Wind Atlas Analysis and Application Program (WAsP). According to the obtained RIX value, the terrain of WPP is found to be complex. Due to the complexity of the terrain, wind power forecasting becomes difficult. To deal with this problem, a forecasting method with fast, accurate, and high performance is needed. Therefore, Extreme Learning Machine (ELM) based method is proposed for wind power forecasting in this study. Electrical and meteorological measurements are obtained from WPP for the application of the proposed method. These measurements are provided with high quality measuring devices. Also, Global Positioning System (GPS) time synchronization is used to prevent lags between measurements. The wind speed, wind direction, and wind power data of 1-year period are obtained from WPP. These data are used to compare the proposed method with a classical Artificial Neural Network (ANN) based method in terms of two, three and four hours-ahead wind power forecast performances. In the forecast studies performed for all data related to 2, 3, and 4-hours ahead, Normalized Root Mean Square Error (NRMSE) values of ELM are obtained as 7.01, 10.12, and 12.06, respectively, while these values are found as 8.19, 12.18, and 13.09 for ANN. In addition, the values of Correlation Coefficients (R) of the proposed forecast method results regarding 2, 3, and 4-hours ahead are 0.96588, 0.93528, and 0.88984, respectively. The R values related to ANN are observed as 0.95421, 0.91373, and 0.87576, respectively. According to the obtained results, it is observed that ELM has better performance features than classic method under all forecast conditions and it is clearly seen that ELM has by far short training time than other one. [ABSTRACT FROM AUTHOR]

Accurate predictions of solar photovoltaic (PV) power generation at different time horizons are essential for reliable operation of energy management systems. The output power of a PV power plant is dependent on non-linear and intermittent environmental factors, such as solar irradiance, wind speed, relative humidity, etc. Intermittency and randomness of solar PV power effect precision of estimation. To address the challenge, this paper presents a Swarm Decomposition Technique (SWD) based hybrid model as a novel approach for very short-term (15 min) solar PV power generation forecast. The original contribution of the study is to investigate use of SWD for solar data forecast. The solar PV power generation data with hourly resolution obtained from the field (grid connected, 857.08 kWp Akgul Solar PV Power Plant in Turkey) are used to develop and validate the forecast model. Specifically, the analysis showed that the hybrid model with SWD technique provides highly accurate predictions in cloudy periods. [ABSTRACT FROM AUTHOR]

The reference price is the most important signal for market participant to have position of selling or buying in the trade of electricity. The electricity price has a dynamic structure and is directly and/or indirectly affected by many factors. All of the market participants' transaction is carried out in a contract based on forecasting. These transactions are for day-ahead market (DAM) and balancing power market 1 day before, intraday market 60 min before and short, mid, long term for derivatives market. The price is forecasted that is determined as far as cost. The price forecasting has benefits which are maximizing profit, protecting from all of the crisis and constrictions and mitigation of loses. All of the market participants highly affected from the price. For this reason, such an important topic should be conducted scientifically. This study aims to develop a price forecasting tool for short term which is calculated day by day in DAM at Turkish electricity market. This forecasting tool is developed based on artificial neural networks. All of the market participants take the price, which is calculated in DAM, as reference. There are many direct and/or indirect factors affecting the reference price used by market players. In this study, a short-term price forecasting model was formed by analyzing historical data and all of the data, which are affecting the price most correctly, are correlated by using artificial neural network method. The basis of the forecasting model used is supply and demand curves. All the factors affecting different points on supply and demand curves by artificial neural network model provide the most correct reference price. In the study, January, April, July and October 2017 values were taken and estimated for randomly selected days of the months. The randomly selected supply and demand curves of April 19 is affected from renewable energy production, natural gas and coal production. In this scope of conclusion, the forecasted price is below 5% deviation on average per day compared to the actual price. As for hourly error deviation, it was seen that deviation is below 10% for 13 h. The suggested artificial neural network model scientifically supports market participant to make short-term transaction. [ABSTRACT FROM AUTHOR]

Turkey's economic development in terms of urbanization, consumption of electricity is a significant pressure on the natural environment. In order to reduce the effects of this pressure in parallel with economic growth, it is necessary to reduce the dependence on foreign sources in energy while reducing the amount of carbon emissions that cause environmental pollution. Therefore, Turkey's economic growth of urbanization in order to continue in a healthy way, the relationship between the amount of energy consumption and carbon emissions need to be better understood. The target to reduce the economic and environmental impacts of economic growth and urbanization with increasing energy needs to the minimum should be to use renewable energy sources efficiently. The aim of this study urbanization in Turkey, the relationship between electricity consumption and carbon emissions is to examine the amount of time series analysis. The increase in the amount of carbon emission, the increase in population and the increase in electricity demand in comparison with years and sectoral basis were compared. discussing the results obtained from the graph are interpreted in how it has changed that and more weight which results results obtained from the graphic industry where the bet is causing the increase of electricity demand with urbanization between 1990-2017 years in Turkey. Graphic shows the results obtained from the last 27 years the amount of carbon emissions in Turkey, the population and electricity demand is increasing in parallel with each other. Especially the rise in demand for electricity and carbon emissions, but not appear as a result of this parallel Turkey's recent emphasis on renewable energy sources. [ABSTRACT FROM AUTHOR]

Solar electricity generation has been investigated in three different sectors as residential, commercial and industrial. This study aims to forecast the solar electricity generation in 5 years by using statistical analysis and make projection of predictions about what could the solar technology can become in Turkey, using USA data on solar electricity generation since statistical analysis is made by using evaluation of regression analysis for these sectors. The results show that the PV generation for the sectors by years such as distributed solar photovoltaic generation value varies from 10.41 MkW h (1989) to 10,464.61 MkW h (2016) for the residential sector; from 14.72 MkW h (1989) to 7179.62 MkW h (2016) for the commercial sector; and from 3.26 MkW h (1989) to 1822.60 MkW h (2016) for the industrial sector. Additionally, forecast of energy distribution (million kW h) has been carried out that from 12783.96 to 28738.31; from 6524.67 to 8018.74; from 1927.91 to 3435.58 with respect to sectors of residential, commercial and industrial, respectively. These results are obtained with the R2 values (0.968667; 0.932558; 0.716024). Consequently, it would not be wrong to express that the solar energy is no more a novelty, but a necessity with the daily electricity usage. In this study, the novelty lies in study's unique combination of analysing goals with a benchmarking evaluation between USA and Turkey. This is illustrated with the forecasting energy distribution of over 8%, a component usage of at least three equivalent sectors. [ABSTRACT FROM AUTHOR]

In deregulated electricity markets the independent system operator (ISO) oversees the power system and manages the supply and demand balancing process. In a typical day the ISO announces the electricity demand forecast for the next day and gives participants an option to prepare offers to meet the demand. In order to have a reliable power system and successful market operation, it is crucial to estimate the electricity demand accurately. In this paper, we develop an hourly demand forecasting method on annual, weekly and daily horizons, using a linear model that takes into account the harmonics of these variations and the modulation of diurnal periodic variations by seasonal variations. The electricity demand exhibits cyclic behavior with different seasonal characteristics. Our model is based solely on sinusoidal variations and predicts hourly variations, without using any climatic or econometric information. The method is applied to the Turkish power market on data for the period 2012–2014 and predicts the demand over daily and weekly horizons within a 3% error margin in the Mean Absolute Percentage Error (MAPE) norm. We also discuss the week day/weekend/holiday consumption profiles to infer the proportion of industrial and domestic electricity consumption. [ABSTRACT FROM AUTHOR]

• Pandemics severely affect the national economies in the short and long terms. • Examines the impact of lockdowns on electricity and natural gas consumption in industrial zones. • Lockdowns during the pandemic was caused a significant decrease in energy consumptions. • Electricity and natural gas consumption amount prediction performance of the ARIMA was better than the Holt-Winters. • COVID-19 lockdowns negatively affect the model prediction accuracy. The COVID-19 lockdowns have adversely affected the national economies and caused fluctuations in the energy industry. This study examined how the lockdowns during the COVID-19 pandemic affected the amount of electricity and natural gas consumption in four organized industrial zones in the Turkey. A significant decrease was observed in electricity and natural gas consumption amounts in April and May when lockdowns were also applied in four industrial zones. In April, electricity consumption decreased between 72 and 43%, and natural gas consumption decreased between 77 and 57%. In May, electricity consumption decreased between 60 and 32%, and natural gas consumption decreased between 69 and 45%. These decreases in industrial zones show that the economy has been significantly affected. Furthermore, in this study, Auto-Regressive Integrated Moving Average (ARIMA) and Holt-Winters models were developed to predict electricity and natural gas consumption of an industrial zone. ARIMA(0,0,2)(2,1,0) 7 and ARIMA(0,0,2)(0,1,1) 7 models were chosen as the best model for the electricity and natural gas consumption data respectively with a minimum MAPE Electricity was 1.37%, RMSE Electricity was 87.2, R2 Electricity was 0.99, MAPE Gas was 5.42% and RMSE Gas was 50.9, R2 Gas was 0.92. Electricity and natural gas consumption was forecasted for the next ten days (10–19 March 2021) according to ARIMA models with 80% and 95% confidence intervals. In addition, in this study, the impact of low energy usage in the industrial zone due to the COVID-19 lockdowns on model prediction performance was also examined. The obtained results showed that the COVID-19 lockdowns were reduced the ARIMA model prediction accuracy. [ABSTRACT FROM AUTHOR]

In the last few years, several countries have accomplished their determined renewable energy targets to achieve their future energy requirements with the foremost aim to encourage sustainable growth with reduced emissions, mainly through the implementation of wind and solar energy. In the present study, we propose and compare five optimized robust regression machine learning methods, namely, random forest, gradient boosting machine (GBM), k-nearest neighbor (kNN), decision-tree, and extra tree regression, which are applied to improve the forecasting accuracy of short-term wind energy generation in the Turkish wind farms, situated in the west of Turkey, on the basis of a historic data of the wind speed and direction. Polar diagrams are plotted and the impacts of input variables such as the wind speed and direction on the wind energy generation are examined. Scatter curves depicting relationships between the wind speed and the produced turbine power are plotted for all of the methods and the predicted average wind power is compared with the real average power from the turbine with the help of the plotted error curves. The results demonstrate the superior forecasting performance of the algorithm incorporating gradient boosting machine regression. [ABSTRACT FROM AUTHOR]

• A novel residual-based convolutional neural network model is developed for efficient wind power forecasting. • Variational mode decomposition contributes significantly to the forecast performance of the network. • The proposed network has both less complexity and less computational cost. • The proposed model provides superior short-term wind power forecasting performance. • The effectiveness of the method is compared with state-of-the-arts pre-trained networks. An accurate forecast of wind power is very important in terms of economic dispatch and the operation of power systems. However, effectively mitigating the risks arising from wind power in power system operations greatly reduces the risk of wind energy producers, exposing them to potential additional costs. Being aware of this challenge, we introduced a two-step novel deep learning method for wind power forecasting. The first stage includes processes of Variational Mode Decomposition (VMD)-based feature extraction and converting these features into images. In the second stage, an improved residual-based deep Convolutional Neural Network (CNN) was utilized to forecast wind power. Meteorological wind speed, wind direction, and wind power data, which are directly related to each other, were employed as a dataset. The combined dataset was procured from a wind farm in Turkey between January 1 and December 31, 2018. The results of the proposed method were compared with the results obtained from the state-of-the-art deep learning architectures namely SqueezeNet, GoogLeNet, ResNet-18, AlexNet, and VGG-16 as well as physical model based on available meteorological forecast data. The proposed method outperformed the other architectures and demonstrated promising results for very short-term wind power forecasting due to its competitive performance. [ABSTRACT FROM AUTHOR]

This study starts with a brief analysis of the global electricity market with special reference to the Turkish market. Next, the daily electricity consumption amounts in Turkey between 2012 and 2016 are examined by using statistical tools. Furthermore, the periodicity in the data is discovered. The periodicity, which has an impact not only in the electricity market but also in other sectors of the economy, is an important indicator for planning and policy-making. Periodicity is easier to observe depending on seasonal impacts. However, what is important is to detect hidden periodicity. The main contribution of this study is to detect the hidden periodicity with a rather novel approach in the electricity industry. The proposed periodicity test based on the periodogram ordinate has two major advantages. First, the periodograms are invariant to the model selection. Secondly, the distributions of the normalized periodograms and therefore critical values do not depend on the sample size. Moreover, the forecasting performance of the model for Turkish electricity consumption seems to be better compared to the standard time series model. • Calculation of forecasting values by applying the harmonic regression method to daily data. • Revealing the hidden periodicities in the data. • Comparison of the harmonic regression model and the ARIMA model. [ABSTRACT FROM AUTHOR]