Battery energy storage system fault state self-adaptive early warning method

Abstract

The invention discloses a battery energy storage system fault state self-adaptive early warning method, which comprises the following steps: carrying out dimension reduction on high-dimensional monitoring data by using an NJW clustering algorithm, obtaining a feature vector by constructing a pull matrix of heterogenous data, and carrying out clustering analysis by using the feature vector to uniquely replace original data, the problem of singularity frequently occurring in non-convex data clustering in a traditional method is solved; using the DTW for dynamically normalizing asynchronous monitoring data time axes, mapping two groups of monitoring data to a synchronous time axis, thus observation errors caused by asynchronous sampling of the monitoring data are overcome, and the problem that the time axes of different-source data of sampling points cannot be in one-to-one correspondence in a traditional method is solved; and finally, constructing a sliding window model to suppress the influence of a small number of outliers in monitoring data, carrying out clustering analysis based on a DTM distance, and objectively selecting a fault clustering point through a sparse coefficient LSR and a fault threshold FT, thereby avoiding over-estimation or under-estimation of a fault state by a traditional fault clustering method, and realizing BESS fault state adaptive early warning.

Description

technical field [0001] The invention relates to the technical field of detection of a battery energy storage system, in particular to an adaptive early warning method for a fault state of a battery energy storage system. Background technique [0002] Most of the safety problems encountered by the battery energy storage system (BESS) come from the cell level, mainly including overcharge, overdischarge, internal short circuit and external short circuit. In the past BESS fault warning, whether it is The parameter estimation method implemented by establishing an accurate electrothermal simulation model or the threshold limit method based on the empirical estimation method only identify abnormal data during the operation of the BESS, and rely heavily on normal data for reference correction, while ignoring normal batteries Observation errors and process noise in cyclic processes. However, large-capacity battery energy storage equipment is generally composed of hundreds of cells c...

AI Technical Summary

Problems solved by technology

[0012] In view of the above problems, the purpose of the present invention is to provide a battery energy storage system self-adaptive early warning method for fault state, based on the NJW spectral clustering algorithm to process heterogeneous data, and to group battery boxes with similar working conditions and fault state comparisons. Class, use dynamic regularization algorithm to solve the problem of inconsistent time axis of asynchronous data, build a sliding window model to suppress the influence of a small number of outliers in monitoring data, perform cluster analysis based on DTW distance, and objectively select fault clusters through sparse coefficient and fault threshold points to realize self-adaptive early warning of BESS fault status

Images