With the continuous increase of wind power penetration in power systems, the problems caused by the volatile nature of wind speed and its occurrence in the system operations such as scheduling and dispatching have drawn attention of system operators, utilities and researchers towards the state-of-the-art wind speed and power forecasting methods These methods have the required capability of reducing the influence of the intermittent wind power on system operations as well as of harvesting the wind energy effectively In this context, combining different methodologies in order to circumvent the challenging model selection and take advantage of the unique strength of plausible models have recently emerged as a promising research area Therefore, a comprehensive research about the combined models is called on for how these models are constructed and affect the forecasting performance Aiming to fill the mentioned research gap, this paper outlines the combined forecasting approaches and presents an up-to date annotated bibliography of the wind forecasting literature Furthermore, the paper also points out the possible further research directions of combined techniques so as to help the researchers in the field develop more effective wind speed and power forecasting methods

A review of combined approaches for prediction of short-term wind speed and power

In this paper, an overview of new and current developments in wind forecasting is given where the focus lies upon principles and practical implementations High penetration of wind power in the electricity system provides many challenges to the power system operators, mainly due to the unpredictability and variability of wind power generation Although wind energy may not be dispatched, an accurate forecasting method of wind speed and power generation can help the power system operators reduce the risk of unreliability of electricity supply This paper gives a literature survey on the categories and major methods of wind forecasting Based on the assessment of wind speed and power forecasting methods, the future development direction of wind forecasting is proposed

https://file.scirp.org/pdf/JPEE_2014041616192284.pdf

A Literature Review of Wind Forecasting Methods

Wind power forecasting based on daily wind speed data using machine learning algorithms

Quantification of uncertainties associated with wind power generation forecasts is essential for optimal management of wind farms and their successful integration into power systems. This paper investigates two neural network-based methods for direct and rapid construction of prediction intervals (PIs) for short-term forecasting of power generation in wind farms. The lower upper bound estimation and bootstrap methods are used to quantify uncertainties associated with forecasts. The effectiveness and efficiency of these two general methods for uncertainty quantification is examined using twenty four month data from a wind farm in Australia. PIs with a confidence level of 90% are constructed for four forecasting horizons: five, ten, fifteen, and thirty minutes. Quantitative measures are applied for objective evaluation and unbiased comparison of PI quality. Demonstrated results indicate that reliable PIs can be constructed in a short time without resorting to complicate computational methods or models. Also quantitative comparison reveals that bootstrap PIs are more suitable for short prediction horizon, and lower upper bound estimation PIs are more appropriate for longer forecasting horizons.

Prediction Intervals for Short-Term Wind Farm Power Generation Forecasts

Short-term wind power forecasting using adaptive neuro-fuzzy inference system combined with evolutionary particle swarm optimization, wavelet transform and mutual information

A Review of Wind Power Forecasting Models

wind turbine SCADA (Supervisory control and data acquisition) data contains a considerable proportion reflecting abnormal operation. Accurately eliminating these abnormal data is the basis for subsequent wind turbine power prediction and generation performance evaluation. This paper proposes a method for cleaning abnormal operation data of wind turbines based on average confidence interval. First of all, all wind turbine operational data are allocated into different bins at a certain interval in the horizontal power direction, and the adaptive kernel density method is used to successively establish the probability density distribution of operation data in each horizontal power bin. Next, the distribution characteristics of abnormal data points are obtained by analyzing the shape of probability density curve. Finally, an average confidence interval method is used to eliminate abnormal data. With actual wind turbine SCADA data, the proposed method is proved to be effective.

Abnormal Data Detection and Cleaning of Wind Turbines Based on Average Confidence Interval Method

A large amount of abnormal data will be generated during the actual wind turbine operation, thus the raw data can’t be directly applied to the subsequent work such as wind turbine power prediction and generation performance evaluation. This paper proposes an abnormal data identification method based on the Dirichlet Process Gaussian Mixture Model (DPGMM) to preprocess the raw data effectively. Firstly, all data points are allocated into corresponding power bins created in the horizontal power direction with a certain interval in the wind speed-power (V-P) coordinate system. And then, the DPGMM model that can adaptively determine the optimal number of Gaussian components is used to cluster the data points in each power bin. At last, combined with the parameters of each Gaussian component confidence ellipse and data points distribution characteristics in V-P coordinate system, the abnormal Gaussian components and their clustering, abnormal data can be accurately identified. Using actual wind turbine SCADA data, the proposed method is demonstrated to be effective.

Abnormal Wind Turbine Data Identification Using a Dirichlet Process Gaussian Mixture Model

The multi-dimensional comprehensive evaluation of relay protection is the refinement of the comprehensive evaluation in the evaluation dimension, which can further analyze the combined characteristics of the operation status of relay protection in multiple dimensions. The multi-dimensional comprehensive evaluation of relay protection is easy to fall into the scene of small sample evaluation, thus affecting the accuracy of the comprehensive evaluation. In order to solve the small sample problem of multi-dimensional comprehensive evaluation of relay protection, this paper proposes the concept of comprehensive index of comprehensive evaluation, which reflects the influence characteristics of each single dimension of comprehensive evaluation of relay protection on the results of multi-dimensional comprehensive evaluation. The multi-dimensional comprehensive evaluation method of relay protection based on single-dimensional evaluation results is put forward accordingly, which can refine the operation state of relay protection under the combined classification, thus completing the refinement and practicality of relay protection operation evaluation. The effectiveness of the method in this paper is verified by an example.

Indexes and Methods of Multi-dimensional Comprehensive Evaluation of Relay Protection

The identification of wind turbine abnormal data is the basis for subsequent wind power prediction and health assessment of wind turbine. This paper analyzes the characteristics of wind turbine abnormal data in detail, and designs the MKIF (Mini batch K Means-Isolation Forest) algorithm to realize the identification of abnormal data of wind turbine. The MKIF algorithm introduces Mini batch K Means clustering into the partition process of the Isolation Forest search tree, and use Silhouette Coefficient to supervise the number and location of the split nodes of the tree. The MKIF anomaly score is defined to describe the degree of isolation of the data, which can effectively identify and eliminate abnormal data and use normal data to establish the main power band. Taking the actual operating data of two wind turbines as an example, the effectiveness of the MKIF abnormal data identification algorithm is verified.

Peng Guo

Papers

A Review of Wind Power Forecasting Models

Abnormal Data Detection and Cleaning of Wind Turbines Based on Average Confidence Interval Method

Abnormal Wind Turbine Data Identification Using a Dirichlet Process Gaussian Mixture Model

Indexes and Methods of Multi-dimensional Comprehensive Evaluation of Relay Protection

Wind Turbine Abnormal Data Identification Based on MKIF Model