Models for monitoring wind farm power
Introduction
The generation of wind energy on an industrial scale is relatively new. It is then natural that the performance of wind power farms has not been adequately studied. One of the weakest points in wind power generation is the low predictive accuracy of the energy output. Like industrial corporations managed by enterprise-wide systems, a software solution for prediction of wind farm performance (including the amount of energy produced) is needed. The envisioned wind farm performance prediction models should be able to predict the amount of energy produced on different time scales, e.g., 10 min, 1 h, a day, etc. Such models could transform a wind farm into a wind power plant.
Researchers have applied different methodologies in studying wind farms. Cameron and Michael [3] combined the fuzzy set and neural network approaches in an adaptive-neurons-fuzzy inference system to forecast a wind time series. Landberg [6] builts a model to predict the power produced by a wind farm using the data from the weather prediction model (HIRLAM) and the local weather model (WASP). Li et al. [13] compared regression and neural network (NN) models in order to estimate a turbine's power curve. They reported that the NN model outperformed the regression model. Goh et al. [11] proposed a neural network architecture, the complex-valued pipelined recurrent neural network (CPRNN) using a complex value (combined wind speed and direction into one complex value) as input, for predicting the turbine output. Santoso and Le [14] focused on modeling fixed-speed wind turbines. They modeled the component blocks of a turbine (for aerodynamic, mechanical, and electrical components), and aggregated them into models of a single turbine and a wind farm. Lange and Focken [28] presented various models for short-term wind power prediction, including physics-based, fuzzy, and neuro-fuzzy models. Barbounis et al. [29] constructed a local recurrent neural network model for long-term wind speed and power forecasting based on the meteorological data. Hourly forecasts for up to 72 h ahead were produced for a wind park.
Wind energy has become one of the most important sources of energy. Building accurate models for predicting power output and health monitoring of wind farms is needed by this new industry. Developing such models is challenging, as a large number of parameters are involved. The high dimensional and stochastic nature of a wind farm environment calls for new modeling approaches. The developments in data mining (DM) and evolutionary computation (EC) offer promising approaches to model wind farms. Numerous applications of data mining in manufacturing, marketing, medical informatics and the energy industry have proven to be effective in support of decision making [2], [5], [7], [8], [24]. Successful applications of evolutionary computation have also been reported in many other domains [1], [4], [9], [10], [12], [15].
In this paper, a variety of different approaches, including data mining, evolutionary computation, principal component analysis (PCA), residual approach, and control charts, have been used to build prediction models and characterize power curves of a wind farm by a nonlinear parametric model. The models are built using historical data collected by SCADA (Supervisory Control and Data Acquisition) systems at a wind farm.
Section snippets
Data description
The data used in this research was generated at a wind farm with about 100 turbines. The data was collected by a SCADA system installed at each wind turbine. Each SCADA system collects data on more than 120 parameters. Though the data is sampled at a high frequency, e.g., 2 s, the data is averaged and stored at 10-min intervals (referred to as 10-min data). The data used in this research was collected over a period of one month at all turbines of the wind farm. The data included one file for
Analysis of outlier data
In this section the 70 points with large relative error (larger than 15%) predicted by the k-NN-P1 (k = 250) model are analyzed.
According to the manual of the wind turbine on the farm, the cut-off wind speed of a single turbine is 3.5 m/s. The data in Table 5 indicates that the average wind speed for all outliers is around 2.4 m/s (below the turbine's cut-off wind speed), and the corresponding average power of 70 outlier data produced is approximately 1010 kW. Therefore, generally, if the average
Nonlinear parametric modeling of wind farm power curves
The quality of the power generated by a wind farm is characterized by its power curve (see Fig. 2). Thus far the existing wind energy literature and practices assume that the power curve is static. This research shows that the power curve is not static, and it should be constructed as parametric. A parametric power curve adjusts to the current operational conditions by modifying its parameters, resulting in an enhanced performance of a wind turbine. One way of using the parametric power curve
Filtering outliers by residual approach and control charts
A formal approach to detect outliers in data is needed. The high quality data (without noise and outliers) can be used to build models of high prediction accuracy when the wind farm operates under normal conditions.
Conclusion
Models for computing power produced by a wind farm under normal operating conditions were developed. In particular, a nonlinear parametric model of a power curve was constructed. To develop these models algorithms from four different domains were used, namely: data mining, evolutionary computation, principal component analysis, and statistical process control. The focus of the paper was on studying a wind farm operating in normal conditions. The normal conditions exclude states where the wind
Acknowledgement
The research reported in the paper has been partially supported by funding from the Iowa Energy Center Grant No. 07-01.
References (29)
- et al.
Parameter estimation of three-winding transformers using genetic algorithm
Engineering Applications of Artificial Intelligence
(2002) - et al.
Fundamental time–domain wind turbine models for wind power studies
Renewable Energy
(2007) - et al.
Multilayer perceptron tumor diagnosis based on chromatography analysis of urinary nucleoside
Neural Networks
(2007) - et al.
Bundling classifiers by bagging trees
Computational Statistics and Data Analysis
(2005) - et al.
Cancer gene search with data-mining and genetic algorithms
Computers in Biology and Medicine
(2007) - et al.
Introduction to evolutionary computation
(2003) - et al.
Combustion efficiency optimization and virtual testing: a data-mining approach
IEEE Transactions on Industrial Informatics
(2006) - et al.
Very short-term wind forecasting for Tasmanian power generation
IEEE Transactions on Power Systems
(2006) - et al.
Optimal design of horizontal-axis wind turbines using blade-element theory and evolutionary computation
Journal of Solar Energy Engineering
(2002) - et al.
Data mining in manufacturing: a review
ASME Transactions: Journal of Manufacturing Science and Engineering
(2006)
Short-term prediction of the power production from wind farms
Journal of Wind Engineering and Industrial Aerodynamics
Data mining techniques: for marketing, sales, and customer relationship management
Introduction to data mining
Placement of wind turbines using genetic algorithm
Renewable Energy
Cited by (218)
A new method for estimating the annual energy production of wind turbines in hot environments
2024, Renewable and Sustainable Energy Reviews1D Convolutional LSTM-based wind power prediction integrated with PkNN data imputation technique
2023, Journal of King Saud University - Computer and Information SciencesA trend-based method for the prediction of offshore wind power ramp
2023, Renewable EnergyA comprehensive review on enhancing wind turbine applications with advanced SCADA data analytics and practical insights
2024, IET Renewable Power Generation