The Anomaly Detection Method for New Energy Vehicle Batteries Based on Conv-Attention-MLP
In terms of the detection of abnormal signals in new energy vehicle batteries,in order to solve the issues that traditional anomaly detection methods have low accuracy and poor generalization in the detection of multi-dimensional battery data,this paper proposes a Conv-Attention-MLP deep neural network method which inte-grates convolution(Conv),the self-attention mechanism(Attention)and the multilayer perceptron(MLP).The method first uses feature convolution to reduce the dimensionality of multidimensional time-series data,then uses the multihead self-attention(MSA)mechanism to recmodel the asynchrony of data time and the interrelationships among data logic,and finally uses the continuous linear mapping and non-linear operation of the multilayer percep-tron(MLP)to generate complex piecewise mapping,so as to achieve better fitting results,which realizes the effec-tive detection of multiple-dimensional battery data,and enhances the model's accuracy and robustness for diverse data.Experimental results demonstrate that compared to traditional anomaly detection methods,the the Conv-Attention-MLP model has better performance on the vloongs dataset,showcasing higher precision and stronger ro-bustness.
Deep learningAnomaly detectionSelf-attention mechanismModel merging
NETL
NSTL
万方数据
