Abstract
A abnormal data repairing method based on adjacent power plant and integrated similar days of BP neural network is presented. Some factors influencing power generation such as geographical position, temperature and day type are considered. By means of selecting adjacent power plants with high power correlation to plant to be repaired by Pearson product-moment correlation coefficient and the combination of Grey Relational Analysis and curve similarity is used to select similar days. Finding out the integrated similar days' data of the adjacent power plant that is in conformity with the day to be repaired, corresponding BP neural network model is built, then the diverse learning speed algorithm are employed to repair abnormal data. The actual abnormal data in PV prediction power repairing results for Qinghai district show that the proposed method possesses better repairing accuracy.
Export citation and abstract BibTeX RIS
Content from this work may be used under the terms of the Creative Commons Attribution 3.0 licence. Any further distribution of this work must maintain attribution to the author(s) and the title of the work, journal citation and DOI.