Unsupervised Anomaly Detection Approach Based on Adversarial Memory Autoencoders for Multivariate Time Series

The widespread usage of Cyber Physical Systems(CPSs)generates a vast volume of time series data,and precisely determining anomalies in the data is critical for practical production.Autoencoder is the mainstream method for time series anomaly detection,and the anomaly is judged by reconstruction error.However,due to the strong generalization ability of neural networks,some abnormal samples close to normal samples may be judged as normal,which fails to detect the abnormality.In addition,the dataset rarely provides sufficient anomaly labels.This research proposes an unsupervised anomaly detection approach based on adversarial memory autoencoders for multivariate time series to solve the above problem.Firstly,an encoder encodes the input data into low-dimensional space to acquire a feature vector.Then,a memory module is used to learn the feature vector’s prototype patterns and update the feature vectors.The updating process allows partial forgetting of information to prevent model overgeneralization.After that,two decoders reconstruct the input data.Finally,this research uses the Peak Over Threshold(POT)method to calculate the threshold to determine anomalous samples from normal samples.This research uses a two-stage adversarial training strategy during model training to enlarge the gap between the reconstruction error of normal and abnormal samples.The proposed method achieves significant anomaly detection results on synthetic and real datasets from power systems,water treatment plants,and computer clusters.The F1 score reached an average of 0.9196 on the five datasets,which is 0.0769 higher than the best baseline method.

Elastic Model of Flexure-Based Linkage under Motion Constrains and External Load

For a flexible mechanism with several-stage flexible linkage, the flexible linkage is equivalent to work under the actions of an external load and motion constrains. This paper aims to deal with a simplified elastic model on the kinematic characteristics of a flexure-based linkage under these conditions. The elastic modeling method was developed based on motion constrains and the elastic beam theorem(EBT). Effects of a constant force, an elastic force with a constant stiffness, and the materials were taken into account. The proposed modeling method was verified by comparing with the finite element method(FEM). Further, the developed modeling method was used to optimize a flexure-based mechanism based on a two-stage flexible linkage to realize a maximum displacement amplification ratio of 6.56. The flexure-based mechanism was employed to drive a miniature sucker, which performed with a negative pressure of 2.45 kPa at a frequency of 13.2 kHz.

BcBRC1a is a negative regulator for tillering in non-heading Chinese cabbage

Shoot branching is a decisive factor for crop yield.Molecular mechanism for regulating shoot branching(tillering)needs to be determined.Plenty of previous studies have illustrated that BRANCHED1(BRC1)is a key integrator of shoot branching regulating signals.However,BcBRC1 function in non-heading Chinese cabbage(Brassica campestris ssp.chinensis)(NHCC)remains unknown.Here,we defined two BRC1 orthologs,BcBRC1a and BcBRC1b,from NHCC and focused on the BcBRC1a gene to describe its alternative splicing characteristic and structure.BcBRC1a was expressed rhythmically and mainly in leaf axils at the'Maertou'cultivar tillering stage.BcBRC1aL encoded a nuclear location protein.Its ectopic expression caused Arabidopsis growth inhibition and silencing BcBRC1a led to increased tiller numbers in'Maertou'.Removing the shoot tips of NHCC caused axillary buds to be released from apical dominance and BcBRC1a expression down-regulation.Our research determined that BcBRC1a acts as a negative regulator for tillering in non-heading Chinese cabbage and sets the foundation for further studies.