Adaptive detection and correction method for anomalous wind speed

To improve the accuracy and availability of the data acquisition system from wind farms, this study proposes to use the adaptive detection methods for effective detection abnormal wind speed. In allusion to characterization of the abnormal value was not obvious, choose Auto Regressive Integrated Moving Average to predict wind value of the current moment to obtain residual sequence. In order to reduce the interference of systematic errors, Using Empirical Mode Decomposition method gets the residual sequence of gross error characteristic information. With dual stochastic process by using Hidden Markov Mode of adaptive detection and removed abnormal value, to avoid the shortcomings of traditional threshold identification methods. Finally, using the cubic spline interpolation correcting abnormal data to get a complete wind speed sequence. RBF forecast results show that paper method to be better than traditional wavelet method and can be improved forecasting accuracy of short-term in wind-speed and power.

[1]  Pierluigi Siano,et al.  A Novel RBF Training Algorithm for Short-Term Electric Load Forecasting and Comparative Studies , 2015, IEEE Transactions on Industrial Electronics.

[2]  Enrico Zio,et al.  An Interval-Valued Neural Network Approach for Uncertainty Quantification in Short-Term Wind Speed Prediction , 2015, IEEE Transactions on Neural Networks and Learning Systems.

[3]  Yan Yonglon.  A Wind Turbine Anomaly Detection Method Based on Information Entropy and Combination Model , 2015 .

[4]  Zhu Qianwe.  Methods for elimination and reconstruction of abnormal power data in wind farms , 2015 .

[5]  Fang-Tao Li,et al.  Wind speed short-term forecast for wind farms based on ARIMA model , 2014 .

[6]  Chellu Chandra Sekhar,et al.  HMM Based Intermediate Matching Kernel for Classification of Sequential Patterns of Speech Using Support Vector Machines , 2013, IEEE Transactions on Audio, Speech, and Language Processing.

[7]  Tang Baiwen.  Application of EMD and Duffing Oscillator to Fault Line Detection in Un-effectively Grounded System , 2013 .

[8]  Ye Lin.  Identification of singular points in wind speed data , 2011 .

[9]  Xu Zhi-gao.  Regression Forecast and Abnormal Data Detection Based on Support Vector Regression , 2009 .

[10]  Wang Jeen-Shing,et al.  A Cluster Validity Measure With Outlier Detection for Support Vector Clustering , 2008, IEEE Transactions on Systems, Man, and Cybernetics, Part B (Cybernetics).

[11]  Y. Ni,et al.  Electricity price forecasting with confidence-interval estimation through an extended ARIMA approach , 2006 .

[12]  Jeff A. Bilmes,et al.  What HMMs Can Do , 2006, IEICE Trans. Inf. Syst..

[13]  H.-L. Lou,et al.  Implementing the Viterbi algorithm , 1995, IEEE Signal Process. Mag..