Performance Evaluation of Primary Frequency Control for Flexible Grid Connected Coal-Fired Units Based on Self-Organizing Maps

Abstract

The primary frequency control (PFC) response performance of coal-fired units is affected by load rate, main steam pressure and grid frequency deviation value, it is difficult to accurately evaluate it online. In this paper, an evaluation method based on k-means clustering analysis and self-organizing map (SOM) is proposed. Firstly, the data segment of primary frequency control is extracted from unit’s operation history data, and the k-means clustering algorithm is used to divide the operational condition space of the data segment. Then, a SOM performance evaluation model using MQE as the evaluation index is established, and the SOM performance evaluation model is trained using the excellent primary frequency modulation performance data set screened in the subspace separately. Finally, for the operating unit, the online extracted primary frequency control data segment is divided into the corresponding working condition subspace according to the principle of the closest distance, and the corresponding trained SOM model is called to obtain the online evaluation value of the unit's primary frequency control performance. It is proved that the method is feasible and effective, and can be used for online evaluation of primary frequency control performance of units, which is of great significance for ensuring safe and reliable operation of power grids.