A Method for Identification of Dominant Parameters of Virtual Load Based on Incremental Learning

Abstract

The invention discloses a virtual load dominant parameter identification method based on incremental learning, including: (1) random value simulation of the virtual load model dominant parameter; (2) establishment of a deep learning neural network; (4) On-line rapid identification and cyclic training; the present invention mainly introduces the feasibility of applying incremental learning to power system analysis, and combines it with load parameter identification to improve training efficiency while ensuring identification accuracy , while preventing catastrophic forgetting while maintaining storage overhead, it provides a new idea for the processing of training samples in parameter identification, and also provides technical support for online identification of dominant parameters of virtual load models; through continuous training online rapid identification The idea is to apply the convolutional neural network to the parameter identification of the load model, and realize the online identification of the dominant parameters of the virtual load model on the big data platform of the power grid, with continuous circulation and continuous training.

Description

technical field [0001] The invention relates to the field of power system load identification, in particular to a virtual load dominant parameter identification method based on incremental learning. Background technique [0002] In the actual power system operation monitoring, accurate load model plays a very important role in the safe and stable operation of the power system. Different load models lead to very different or even completely opposite results for the stability calculation. Therefore, how to establish an accurate load model and obtain accurate model parameters has always been a hot topic for scholars and has been widely concerned for a long time. There are two main types of load modeling methods, namely statistical synthesis method and overall measurement method. The statistical synthesis method first classifies the load, then counts the characteristics of each type of load, and obtains the overall characteristics of the load. The disadvantage of the statisti...

AI Technical Summary

Problems solved by technology

However, in practical applications, it is usually impossible to obtain all the training samples at once, but to obtain them gradually over time, and the information reflected by the samples may also change over time.

If new samples need to relearn all the data after they arrive, it will consume a lot of time and space, so batch learning algorithms cannot meet this demand.

Images