A Fault Prediction Method for Energy Storage Converter

Abstract

The invention discloses a fault prediction method for an energy storage converter, which comprises the following steps: taking the historical monitoring signals of the energy storage converter clusters of the same battery energy storage system as the original feature library, and using a sparse self-encoding algorithm to obtain the original feature library Extract the main feature matrix of the energy storage converter cluster at each sampling time, search for the cluster center energy storage converter at each sampling time based on the fast clustering algorithm, and calculate the cumulative eccentricity distance matrix of the energy storage converter cluster , normalize the cumulative eccentricity distance matrix and set the warning threshold, and finally realize the prediction of energy storage converter failure. The invention realizes the prediction of the failure of the energy storage converter, can run online, is convenient for calculation, has no special requirements and restrictions, is suitable for clusters of energy storage converters of different scales, has good portability, and is beneficial for maintenance personnel to establish a reasonable and effective Maintenance plan to ensure the safe and stable operation of the power grid.

Description

technical field [0001] The invention relates to a fault prediction method for an energy storage converter, which belongs to the technical field of power grids. Background technique [0002] In the smart grid, energy storage has become an important supporting technology for large-scale centralized and distributed new energy generation access and consumption. As a key component of the battery energy storage system, the energy storage converter controls the energy flow between the battery and the grid, and its health status directly affects the safety and stability of the entire battery energy storage system. With the expansion of the application scale of energy storage systems, the power grid has also put forward higher requirements for the health status assessment technology of energy storage converters. Therefore, real-time monitoring of the operating status of the energy storage converter and timely and accurate prediction of the occurrence of energy storage converter fail...

AI Technical Summary

Problems solved by technology

[0003] At present, the maintenance of energy storage converters usually adopts after-the-fact maintenance, and it is difficult for maintenance personnel to grasp the health status of energy storage converters in real time

Fault prediction technology can help maintenance personnel to predict possible faults of energy storage converters in advance. However, most of the existing fault prediction methods rely on the operation data of the entire life cycle of the equipment, and the established fault prediction model is only applicable to a single device. , the portability of the model is poor, and there is still a lack of an effective and scalable fault prediction method for energy storage converters

Images