One class of islanding detection methods, known as impedance measurement-based methods and voltage change monitoring-based methods, are implemented through injecting irregular currents into the network, for which reason they are defined in this paper as irregular current injection methods. This paper indicates that such methods may be affected by distributed generation (DG) unit cut-in events. Although the network impedance change can still be used as a judgment basis for islanding detection, the general impedance measurement scheme cannot separate island events from DG unit cut-in events in multi-DG operation. In view of this, this paper proposes a new islanding detection method based on an improved impedance measurement scheme, i.e., dynamic impedance measurement, which will not be affected by DG unit cut-in events and can further assist some other equipment in islanding detection. The simulations and experiments verify the stated advantages of the new islanding detection method.

Islanding detection methods, based on injecting high-/low-frequency currents or negative sequence fundamental frequency currents and observing the resultant responses, are collectively referred to as irregular current injection methods in this paper. In multi-distributed generation (DG) operation, if there is no restriction to the phase of injected irregular currents, the currents at the same frequency may cancel each other out, and then their convergent current may be too small to cause a detectable response, for which reason islanding detection will be severely affected. Accordingly, this paper raises a compatibility issue, which requires the phase difference between any two injected irregular currents to be within a certain interval. In response to this issue, a solution is proposed. According to this solution, the terminal voltage of DG units is referenced to conduct irregular currents injection, and only certain high-frequency currents are used as injected currents. If this solution is adopted by as many manufacturers as possible, the effect and reliability of such methods will be greatly improved.

Short‐term wind‐power forecasting methods like neural networks are trained by empirical risk minimization. The local optimum and overfitting problem is likely to occur in the model‐training stage, leading to the poor ability of reasoning and generalization in the prediction stage. To solve the problem, a model of short‐term wind power forecasting is proposed based on 2‐stage feature selection and a supervised random forest in the paper. First, in data preprocessing, some redundant features can be removed by a variable importance measure method and intimate samples can be selected based on relevant analysis, so that the efficiency of model training and the correlation degree between input and output samples can be enhanced. Second, an improved supervised random forest (RF) methodology is proposed to compose a new RF based on evaluating the performance of each decision tree and restructuring the decision trees. A new index of external validation in correlation with wind speed in numerical weather prediction has been proposed to overcome the shortcomings of the internal validation index that seriously depends on the training samples. The simulation examples have verified the rationality and feasibility of the improvement. Case studies of measured data from a wind farm have shown that the proposed model has a better performance than the original RF, back propagation neural network, Bayesian network, and support vector machine, in aspects of ensuring accuracy, efficiency, and robustness, and especially if there is high rate of noisy data and wind power curtailment duration in the historical data.

(2018) 'Problems in the classic frequency shift islanding detection methods applied to energy storage converters and a coping strategy.', IEEE transactions on energy conversion., 33 (2). pp. 496-505. Further information on publisher's website: Additional information: Use policyThe full-text may be used and/or reproduced, and given to third parties in any format or medium, without prior permission or charge, for personal research or study, educational, or not-for-prot purposes provided that:• a full bibliographic reference is made to the original source • a link is made to the metadata record in DRO • the full-text is not changed in any way The full-text must not be sold in any format or medium without the formal permission of the copyright holders.Please consult the full DRO policy for further details.