The Experts below are selected from a list of 3471 Experts worldwide ranked by ideXlab platform
Mehmet Yesilbudak - One of the best experts on this subject based on the ideXlab platform.
-
A new approach to very short term wind speed prediction using k-nearest neighbor classification
Energy Conversion and Management, 2013Co-Authors: Mehmet Yesilbudak, Seref Sagiroglu, Ilhami ColakAbstract:Abstract Wind energy is an inexhaustible energy source and wind power production has been growing rapidly in recent years. However, wind power has a non-schedulable nature due to wind speed variations. Hence, wind speed prediction is an indispensable requirement for power system operators. This paper predicts wind speed parameter in an n -tupled inputs using k -nearest neighbor ( k -NN) classification and analyzes the effects of input parameters, nearest neighbors and Distance metrics on wind speed prediction. The k -NN classification model was developed using the object oriented programming techniques and includes Manhattan and Minkowski Distance metrics except from Euclidean Distance metric on the contrary of literature. The k -NN classification model which uses wind direction, air temperature, atmospheric pressure and relative humidity parameters in a 4-tupled space achieved the best wind speed prediction for k = 5 in the Manhattan Distance metric. Differently, the k -NN classification model which uses wind direction, air temperature and atmospheric pressure parameters in a 3-tupled inputs gave the worst wind speed prediction for k = 1 in the Minkowski Distance metric.
-
Very short term pitch angle optimization in wind turbines: A machine learning approach
4th International Conference on Power Engineering Energy and Electrical Drives, 2013Co-Authors: Mehmet Yesilbudak, Seref Sagiroglu, Ersan Kabalci, Ilhami ColakAbstract:This paper proposes a pitch angle forecasting model based on the k-nearest neighbor classification. Air temperature, atmosphere pressure, wind direction, wind speed, rotor speed and wind power parameters were represented as a 6-dimensional attribute tuple in the forecasting model. Euclidean, Manhattan and Minkowski Distance metrics for measuring the proximity between training and test tuples, mean absolute, mean absolute percentage, and normalized root mean square error metrics for measuring the forecasting accuracy were embedded into the forecasting model. The k-nearest neighbor classifier with Manhattan Distance metric for k=1 achieved MAE, MAPE and NRMSE as 0.001°, 0.245% and 0.324%, respectively as the best forecasting accuracy. However, as the worst forecasting accuracy, MAE, MAPE and NRMSE were achieved as 0.015°, 3.236% and 2.613%, respectively for Minkowski Distance metric and k=10.
-
A data mining approach: Analyzing wind speed and insolation period data in Turkey for installations of wind and solar power plants
Energy Conversion and Management, 2013Co-Authors: Ilhami Colak, Seref Sagiroglu, Mehmet Demirtas, Mehmet YesilbudakAbstract:Abstract Wind and solar power plant installations have been recently increased rapidly with respect to the depletion of fossil-based fuels all over the world. Due to stochastic nature of meteorological conditions, wind and solar energies have a non-schedulable nature and they require several installation analyses to determine the location and the capacities of wind and solar power to be produced. This paper focuses on the similarity, feasibility and numerical analyses of 75 cities in Turkey based on the monthly average wind speed and insolation period data. The nearest and the farest neighbor algorithms are used as agglomerative hierarchical clustering methods with Euclidean, Manhattan and Minkowski Distance metrics in the stage of making the similarity and feasibility analyses. The maximum cophenetic correlation coefficient is achieved by the nearest neighbor algorithm with the Minkowski Distance metric in the similarity and feasibility analyses. On the other hand, graphical representations of the monthly average wind speed and insolation period data are utilized for making the numerical analysis. The highest annual average wind speed and insolation period are obtained as 3.88 m/s and 8.45 h/day, respectively. Overall, many inferences were achieved in acceptable and efficient limits for wind and solar energy.
-
ICMLA (2) - Excitation Current Forecasting for Reactive Power Compensation in Synchronous Motors: A Data Mining Approach
2012 11th International Conference on Machine Learning and Applications, 2012Co-Authors: Ramazan Bayindir, Mehmet Yesilbudak, Ilhami Colak, Seref SagirogluAbstract:Excitation current of a synchronous motor has a key role in reactive power compensation. For this purpose, the k-nearest neighbor (k-NN) classifier designed in this paper predicts the excitation current parameter using n-tupled inputs. Load current, power factor, power factor error and the change of excitation current parameters were utilized in n-tupled inputs. Moreover, Euclidean, Manhattan and Minkowski Distance metrics were employed for measuring the closeness among the observations and the nearest neighbor number k was assigned as 1, 2, 3, 4 and 5, respectively. The forecasting results have shown that the k-NN classifier which uses power factor and the change of excitation current parameters achieved the best forecasting accuracy for k=1 in Minkowski Distance metric. However, the k-NN classifier which uses load current, power factor and power factor error parameters gave the worst forecasting accuracy for k=5 in Minkowski Distance metric.
Ilhami Colak - One of the best experts on this subject based on the ideXlab platform.
-
A new approach to very short term wind speed prediction using k-nearest neighbor classification
Energy Conversion and Management, 2013Co-Authors: Mehmet Yesilbudak, Seref Sagiroglu, Ilhami ColakAbstract:Abstract Wind energy is an inexhaustible energy source and wind power production has been growing rapidly in recent years. However, wind power has a non-schedulable nature due to wind speed variations. Hence, wind speed prediction is an indispensable requirement for power system operators. This paper predicts wind speed parameter in an n -tupled inputs using k -nearest neighbor ( k -NN) classification and analyzes the effects of input parameters, nearest neighbors and Distance metrics on wind speed prediction. The k -NN classification model was developed using the object oriented programming techniques and includes Manhattan and Minkowski Distance metrics except from Euclidean Distance metric on the contrary of literature. The k -NN classification model which uses wind direction, air temperature, atmospheric pressure and relative humidity parameters in a 4-tupled space achieved the best wind speed prediction for k = 5 in the Manhattan Distance metric. Differently, the k -NN classification model which uses wind direction, air temperature and atmospheric pressure parameters in a 3-tupled inputs gave the worst wind speed prediction for k = 1 in the Minkowski Distance metric.
-
Very short term pitch angle optimization in wind turbines: A machine learning approach
4th International Conference on Power Engineering Energy and Electrical Drives, 2013Co-Authors: Mehmet Yesilbudak, Seref Sagiroglu, Ersan Kabalci, Ilhami ColakAbstract:This paper proposes a pitch angle forecasting model based on the k-nearest neighbor classification. Air temperature, atmosphere pressure, wind direction, wind speed, rotor speed and wind power parameters were represented as a 6-dimensional attribute tuple in the forecasting model. Euclidean, Manhattan and Minkowski Distance metrics for measuring the proximity between training and test tuples, mean absolute, mean absolute percentage, and normalized root mean square error metrics for measuring the forecasting accuracy were embedded into the forecasting model. The k-nearest neighbor classifier with Manhattan Distance metric for k=1 achieved MAE, MAPE and NRMSE as 0.001°, 0.245% and 0.324%, respectively as the best forecasting accuracy. However, as the worst forecasting accuracy, MAE, MAPE and NRMSE were achieved as 0.015°, 3.236% and 2.613%, respectively for Minkowski Distance metric and k=10.
-
A data mining approach: Analyzing wind speed and insolation period data in Turkey for installations of wind and solar power plants
Energy Conversion and Management, 2013Co-Authors: Ilhami Colak, Seref Sagiroglu, Mehmet Demirtas, Mehmet YesilbudakAbstract:Abstract Wind and solar power plant installations have been recently increased rapidly with respect to the depletion of fossil-based fuels all over the world. Due to stochastic nature of meteorological conditions, wind and solar energies have a non-schedulable nature and they require several installation analyses to determine the location and the capacities of wind and solar power to be produced. This paper focuses on the similarity, feasibility and numerical analyses of 75 cities in Turkey based on the monthly average wind speed and insolation period data. The nearest and the farest neighbor algorithms are used as agglomerative hierarchical clustering methods with Euclidean, Manhattan and Minkowski Distance metrics in the stage of making the similarity and feasibility analyses. The maximum cophenetic correlation coefficient is achieved by the nearest neighbor algorithm with the Minkowski Distance metric in the similarity and feasibility analyses. On the other hand, graphical representations of the monthly average wind speed and insolation period data are utilized for making the numerical analysis. The highest annual average wind speed and insolation period are obtained as 3.88 m/s and 8.45 h/day, respectively. Overall, many inferences were achieved in acceptable and efficient limits for wind and solar energy.
-
ICMLA (2) - Excitation Current Forecasting for Reactive Power Compensation in Synchronous Motors: A Data Mining Approach
2012 11th International Conference on Machine Learning and Applications, 2012Co-Authors: Ramazan Bayindir, Mehmet Yesilbudak, Ilhami Colak, Seref SagirogluAbstract:Excitation current of a synchronous motor has a key role in reactive power compensation. For this purpose, the k-nearest neighbor (k-NN) classifier designed in this paper predicts the excitation current parameter using n-tupled inputs. Load current, power factor, power factor error and the change of excitation current parameters were utilized in n-tupled inputs. Moreover, Euclidean, Manhattan and Minkowski Distance metrics were employed for measuring the closeness among the observations and the nearest neighbor number k was assigned as 1, 2, 3, 4 and 5, respectively. The forecasting results have shown that the k-NN classifier which uses power factor and the change of excitation current parameters achieved the best forecasting accuracy for k=1 in Minkowski Distance metric. However, the k-NN classifier which uses load current, power factor and power factor error parameters gave the worst forecasting accuracy for k=5 in Minkowski Distance metric.
Ersan Kabalci - One of the best experts on this subject based on the ideXlab platform.
-
Very short term pitch angle optimization in wind turbines: A machine learning approach
4th International Conference on Power Engineering Energy and Electrical Drives, 2013Co-Authors: Mehmet Yesilbudak, Seref Sagiroglu, Ersan Kabalci, Ilhami ColakAbstract:This paper proposes a pitch angle forecasting model based on the k-nearest neighbor classification. Air temperature, atmosphere pressure, wind direction, wind speed, rotor speed and wind power parameters were represented as a 6-dimensional attribute tuple in the forecasting model. Euclidean, Manhattan and Minkowski Distance metrics for measuring the proximity between training and test tuples, mean absolute, mean absolute percentage, and normalized root mean square error metrics for measuring the forecasting accuracy were embedded into the forecasting model. The k-nearest neighbor classifier with Manhattan Distance metric for k=1 achieved MAE, MAPE and NRMSE as 0.001°, 0.245% and 0.324%, respectively as the best forecasting accuracy. However, as the worst forecasting accuracy, MAE, MAPE and NRMSE were achieved as 0.015°, 3.236% and 2.613%, respectively for Minkowski Distance metric and k=10.
-
Development of a feasibility prediction tool for solar power plant installation analyses
Applied Energy, 2011Co-Authors: Ersan KabalciAbstract:The solar energy becomes a challenging area among other renewable sources since the solar energy sources have the advantages of not causing pollution, having low maintenance cost, and not producing noise due to the absence of the moving parts. Although these advantages, the installation cost of a solar power plant is considerably high. However, feasibility analyses have a great role before installation in order to determine the most appropriate power plant site. Despite there are many methods used in feasibility analysis, this paper is focused on a new intelligent method based on an agglomerative hierarchical clustering approach. The solar irradiation and insolation parameters of Central Anatolian Region of Turkey are evaluated utilizing the intelligent feasibility analysis tool developed in this study. The clustering operation in the tool is performed by using the nearest neighbor algorithm. At the stage of determining the optimum hierarchical clustering results, Euclidean, Manhattan and Minkowski Distance metrics are adapted to the tool. The achieved clustering results based on Minkowski Distance metric provide the most feasible inferences to knowledge domain expert according to other Distance metrics.
Seref Sagiroglu - One of the best experts on this subject based on the ideXlab platform.
-
A new approach to very short term wind speed prediction using k-nearest neighbor classification
Energy Conversion and Management, 2013Co-Authors: Mehmet Yesilbudak, Seref Sagiroglu, Ilhami ColakAbstract:Abstract Wind energy is an inexhaustible energy source and wind power production has been growing rapidly in recent years. However, wind power has a non-schedulable nature due to wind speed variations. Hence, wind speed prediction is an indispensable requirement for power system operators. This paper predicts wind speed parameter in an n -tupled inputs using k -nearest neighbor ( k -NN) classification and analyzes the effects of input parameters, nearest neighbors and Distance metrics on wind speed prediction. The k -NN classification model was developed using the object oriented programming techniques and includes Manhattan and Minkowski Distance metrics except from Euclidean Distance metric on the contrary of literature. The k -NN classification model which uses wind direction, air temperature, atmospheric pressure and relative humidity parameters in a 4-tupled space achieved the best wind speed prediction for k = 5 in the Manhattan Distance metric. Differently, the k -NN classification model which uses wind direction, air temperature and atmospheric pressure parameters in a 3-tupled inputs gave the worst wind speed prediction for k = 1 in the Minkowski Distance metric.
-
Very short term pitch angle optimization in wind turbines: A machine learning approach
4th International Conference on Power Engineering Energy and Electrical Drives, 2013Co-Authors: Mehmet Yesilbudak, Seref Sagiroglu, Ersan Kabalci, Ilhami ColakAbstract:This paper proposes a pitch angle forecasting model based on the k-nearest neighbor classification. Air temperature, atmosphere pressure, wind direction, wind speed, rotor speed and wind power parameters were represented as a 6-dimensional attribute tuple in the forecasting model. Euclidean, Manhattan and Minkowski Distance metrics for measuring the proximity between training and test tuples, mean absolute, mean absolute percentage, and normalized root mean square error metrics for measuring the forecasting accuracy were embedded into the forecasting model. The k-nearest neighbor classifier with Manhattan Distance metric for k=1 achieved MAE, MAPE and NRMSE as 0.001°, 0.245% and 0.324%, respectively as the best forecasting accuracy. However, as the worst forecasting accuracy, MAE, MAPE and NRMSE were achieved as 0.015°, 3.236% and 2.613%, respectively for Minkowski Distance metric and k=10.
-
A data mining approach: Analyzing wind speed and insolation period data in Turkey for installations of wind and solar power plants
Energy Conversion and Management, 2013Co-Authors: Ilhami Colak, Seref Sagiroglu, Mehmet Demirtas, Mehmet YesilbudakAbstract:Abstract Wind and solar power plant installations have been recently increased rapidly with respect to the depletion of fossil-based fuels all over the world. Due to stochastic nature of meteorological conditions, wind and solar energies have a non-schedulable nature and they require several installation analyses to determine the location and the capacities of wind and solar power to be produced. This paper focuses on the similarity, feasibility and numerical analyses of 75 cities in Turkey based on the monthly average wind speed and insolation period data. The nearest and the farest neighbor algorithms are used as agglomerative hierarchical clustering methods with Euclidean, Manhattan and Minkowski Distance metrics in the stage of making the similarity and feasibility analyses. The maximum cophenetic correlation coefficient is achieved by the nearest neighbor algorithm with the Minkowski Distance metric in the similarity and feasibility analyses. On the other hand, graphical representations of the monthly average wind speed and insolation period data are utilized for making the numerical analysis. The highest annual average wind speed and insolation period are obtained as 3.88 m/s and 8.45 h/day, respectively. Overall, many inferences were achieved in acceptable and efficient limits for wind and solar energy.
-
ICMLA (2) - Excitation Current Forecasting for Reactive Power Compensation in Synchronous Motors: A Data Mining Approach
2012 11th International Conference on Machine Learning and Applications, 2012Co-Authors: Ramazan Bayindir, Mehmet Yesilbudak, Ilhami Colak, Seref SagirogluAbstract:Excitation current of a synchronous motor has a key role in reactive power compensation. For this purpose, the k-nearest neighbor (k-NN) classifier designed in this paper predicts the excitation current parameter using n-tupled inputs. Load current, power factor, power factor error and the change of excitation current parameters were utilized in n-tupled inputs. Moreover, Euclidean, Manhattan and Minkowski Distance metrics were employed for measuring the closeness among the observations and the nearest neighbor number k was assigned as 1, 2, 3, 4 and 5, respectively. The forecasting results have shown that the k-NN classifier which uses power factor and the change of excitation current parameters achieved the best forecasting accuracy for k=1 in Minkowski Distance metric. However, the k-NN classifier which uses load current, power factor and power factor error parameters gave the worst forecasting accuracy for k=5 in Minkowski Distance metric.
Ying Sun - One of the best experts on this subject based on the ideXlab platform.
-
improving robots swarm aggregation performance through the Minkowski Distance function
International Conference on Mechatronics, 2020Co-Authors: Belkacem Khaldi, Fouzi Harrou, Foudil Cherif, Ying SunAbstract:In this work, we study a simple collective behaviour, called aggregation, performed by a swarm of mobile robots system. We mainly proposed the Distance-Minkowski k-Nearest Neighbours (DM-KNN) as a new approach to the aggregation behaviour of simple robots swarm system. The method introduced the Minkowski Distance function in computing Distances between robots' neighbours. In this approach, the set k-nn members with which each robot will interact with is identified. Then an artificial viscoelastic mesh among the set members is built to perform the aggregation. When Analyzing experimental results based on ARGoS, a significant improvement in the aggregation performance of the swarm is shown compared to the classical Distance-weighted k-NN aggregation approach.
-
ICMRE - Improving robots swarm aggregation performance through the Minkowski Distance function
2020 6th International Conference on Mechatronics and Robotics Engineering (ICMRE), 2020Co-Authors: Belkacem Khaldi, Fouzi Harrou, Foudil Cherif, Ying SunAbstract:In this work, we study a simple collective behaviour, called aggregation, performed by a swarm of mobile robots system. We mainly proposed the Distance-Minkowski k-Nearest Neighbours (DM-KNN) as a new approach to the aggregation behaviour of simple robots swarm system. The method introduced the Minkowski Distance function in computing Distances between robots' neighbours. In this approach, the set k-nn members with which each robot will interact with is identified. Then an artificial viscoelastic mesh among the set members is built to perform the aggregation. When Analyzing experimental results based on ARGoS, a significant improvement in the aggregation performance of the swarm is shown compared to the classical Distance-weighted k-NN aggregation approach.