Photovoltaic system depth anomaly detection method based on k-nearest neighbor adaptive voting

Abstract

The invention provides a photovoltaic system depth anomaly detection method based on k-nearest neighbor adaptive voting. The method comprises the following steps: acquiring one-dimensional time seriesdata in a photovoltaic array accumulated in a photovoltaic power generation system, segmenting original data at a fixed length by using a sliding window, converting the original data into a two-dimensional frequency domain data image through fast Fourier transform, classifying external data in combination with slicing time, and giving a label to construct a data set; and training a classificationmodel on the data set by using a deep neural network, and extracting implicit features as anomaly detection input; carrying out data conversion on the input data and extracting hidden layer coding characteristics; in the local hidden layer coding feature set, selecting the voting point for the first time based on the distance measurement by using the double k-neighbors, and obtaining the voting weight for determining whether the voting is abnormal or not adaptively through the distance difference between the voting point and the second k-neighbour point set, so that the method is robust, andthe detection accuracy is improved. The method does not need to compare the real data of the next moment, and improves the real-time detection efficiency.

Description

technical field [0001] The invention relates to a depth anomaly detection method of a photovoltaic system based on k-nearest neighbor adaptive voting, and belongs to the technical field of photovoltaic systems. Background technique [0002] Photovoltaic power generation system is a complex system affected by multiple natural factors such as temperature, irradiance, and smog. It is mainly composed of grid-connected inverters, combiner boxes, and photovoltaic arrays. As a DC power generation unit, the photovoltaic array plays an indispensable role in the normal operation of the entire photovoltaic power generation system. When studying the abnormal detection and fault location of related equipment in photovoltaic systems, on the one hand, various natural factors must be considered, and on the other hand, various types of fault factors within the photovoltaic power generation system must also be considered. [0003] The existing photovoltaic array anomaly detection and early w...

AI Technical Summary

Problems solved by technology

However, the state data in this way is sensitive to the external environment, and the analysis results are easily disturbed by complex natural factors such as temperature, irradiance, and haze. Therefore, there are also defects in the detection accuracy and positioning accuracy.

Images