No abstract

Abstract. Very short-term forecasting of wind power is important to scheduling staff's planning and wind turbine control. This paper has established a combined forecasting model based on Markov chain and support vector machine (SVM). Firstly, the SVM is used to model for wind power. Then, transition probability matrix is made based on Markov chain to modify for SVM prediction. Finally, the prediction confidence interval of combination forecasting model is given by method of fluctuation confidence interval. Verified by an example of a wind farm indicating that the combination forecasting model is better than a single SVM model on a variety of error indicators.

In order to solve the problem of manual input of K value in traditional K-means clustering algorithm, a method to obtain K value automatically is proposed. Firstly, the data needed to be clustered are sampled, and the distance between the data in the sample group is calculated, Then the distance matrix is formed and the de-noising process is done, and the clustering number K is selected based on the principle of distance maximization. Finally, we take two-dimensional data as an example to simulate the K-means clustering algorithm of the number of clusters K based on the principle of distance maximization, then the simulation results are verified by MATLAB. The experimental results show that the algorithm can obtain K value automatically and improve the accuracy of clustering.

No abstract

Abstract-The operation of distribution network will be greatly affected when distributed generation (DG) is accessed to the network. The degree of influence is closely related to the location and capacity of distributed generation. Therefore, it is of great significance to evaluate the consequence of installing distributed generation to distribution network. This research was carried out from the perspective of reactive power optimization, and the evaluation was performed by super-efficiency data envelopment analysis (DEA). The research is expected to reduce the active power loss of distribution network, to reduce the voltage offset and to make the program more economical. The proposed method could objectively sort the different schemes of DG access to distribution network. Finally, an example was given to show that the proposed method can rank the efficiency values of the evaluation scheme and select the optimal scheme, which is feasible in engineering application.