Detection and diagnosis of dynamics in time series data: Theory of noise reduction

1994 ◽  
Author(s):  
Robert Cawley ◽  
Guan-Hson Hsu ◽  
Liming W. Salvino
Keyword(s):  
Time Series ◽  
Noise Reduction ◽  
Time Series Data ◽  
Series Data ◽  
Detection And Diagnosis

Noise reduction in chaotic time series data

Pramana ◽  
1999 ◽  
Vol 52 (1) ◽  
pp. 25-37 ◽  
Author(s):  
A. Bhowal ◽  
T. K. Roy
Keyword(s):  
Time Series ◽  
Noise Reduction ◽  
Time Series Data ◽  
Chaotic Time Series ◽  
Series Data

scholarly journals Relation Inference among Sensor Time Series in Smart Buildings with Metric Learning

2020 ◽  
Vol 34 (04) ◽  
pp. 4683-4690 ◽  
Author(s):  
Shuheng Li ◽  
Dezhi Hong ◽  
Hongning Wang
Keyword(s):  
Time Series ◽  
Demand Response ◽  
Real World ◽  
Time Series Data ◽  
Metric Learning ◽  
Fault Detection And Diagnosis ◽  
Series Data ◽  
Smart Building ◽  
Deep Metric Learning ◽  
Detection And Diagnosis

Smart Building Technologies hold promise for better livability for residents and lower energy footprints. Yet, the rollout of these technologies, from demand response controls to fault detection and diagnosis, significantly lags behind and is impeded by the current practice of manual identification of sensing point relationships, e.g., how equipment is connected or which sensors are co-located in the same space. This manual process is still error-prone, albeit costly and laborious.We study relation inference among sensor time series. Our key insight is that, as equipment is connected or sensors co-locate in the same physical environment, they are affected by the same real-world events, e.g., a fan turning on or a person entering the room, thus exhibiting correlated changes in their time series data. To this end, we develop a deep metric learning solution that first converts the primitive sensor time series to the frequency domain, and then optimizes a representation of sensors that encodes their relations. Built upon the learned representation, our solution pinpoints the relationships among sensors via solving a combinatorial optimization problem. Extensive experiments on real-world buildings demonstrate the effectiveness of our solution.

Multiple-channel Singular Spectrum Analysis-based noise reduction for MT data observed in Boso Peninsula, Japan

2021 ◽  
Author(s):  
Shu Kaneko ◽  
Katsumi Hattori ◽  
Toru Mogi ◽  
Chie Yoshino
Keyword(s):  
Time Series ◽  
Noise Reduction ◽  
Spectrum Analysis ◽  
Reduction Method ◽  
Time Series Data ◽  
Singular Spectrum Analysis ◽  
Series Data ◽  
Boso Peninsula ◽  
Singular Spectrum ◽  
Kanto Earthquake

<p>Off the coast of the Boso Peninsula, there is a triple junction of the Pacific Plate, the Philippine Sea Plate, and the North American Plate and the Boso Peninsula is one of the seismically active areas in Japan. There are also epicenter areas such as the 1703 Genroku Kanto Earthquake (M8.2), the 1923 Taisho Kanto Earthquake (M7.9), and the Boso Slow Slip which occurs every 6 years, which are geologically interesting places. To estimate the subsurface resistivity structure of the whole Boso area, Magnetotelluric (MT) survey with 41 sites (inter-sites distance of 7 km) has been conducted in 2014-2016, using U43 (12 sites, 1 Hz sampling ; Tierra Technica) and MTU-5, 5A, net (41 sites, 15, 150, and 2400 Hz sampling; Phoenix Geophysics). However, the Boso area is greatly affected by leak current from DC-driven trains, factories, and power lines, so the observed data are contaminated by artificial noises. When we tried to apply the conventional noise reduction method (e.g., remote reference (Gamble et al., 1979) and BIRRP (Chave and Thomson, 2004)) in frequency domain, the obtained MT sounding curve was not ideal. In particular, the phase between the periods of 20 and 400 sec was close to 0 degrees. It suggests that the method used is insufficient to reduce the near-field effect for the Boso data. Thus, we developed a new noise reduction method using MSSA (Multi-channel Singular Spectrum Analysis) as a pre-processing method in time domain.</p><p>The procedure is as follows;</p><p>(1) Decompose 6 component data (Hx, Hy, Ex, Ey, Hxr and Hyr: H and E means magnetic and electric field, respectively, x and y indicates NS and EW component, and r denotes the reference field observed at a quiet station) using MSSA into 6×M principal components (PCs).  Here, M shows the window length of MSSA.</p><p>(2) Check contribution and periods of each PC and eliminate the PCs which are corresponding to the longer periods of variation. That is “detrend” of the original data.</p><p>(3) Apply the second MSSA to the detrended time series data to separate signals and noises shorter than 400 sec.</p><p>(4) Calculating correlation coefficients between H and Hr and between E and Hr for each PC and select the PCs with higher correlation to reconstruct time series data to make MT analysis.</p><p>Then, we perform MT analysis by BIRRP to estimate apparent resistivity,</p><p>As a result, the coherences of H-Hr, and E-Hr were improved and the MT sounding curve became smoother than those results by the conventional noise reduction methods. This indicated that the effectiveness of the proposed noise reduction. However, further investigation in different periods and sites will be required.</p>

scholarly journals Noise Reduction for Nonlinear Nonstationary Time Series Data using Averaging Intrinsic Mode Function

Algorithms ◽  
2013 ◽  
Vol 6 (3) ◽  
pp. 407-429 ◽  
Author(s):  
Bhusana Premanode ◽  
Jumlong Vongprasert ◽  
Christofer Toumazou
Keyword(s):  
Time Series ◽  
Noise Reduction ◽  
Time Series Data ◽  
Intrinsic Mode Function ◽  
Nonstationary Time Series ◽  
Series Data ◽  
Mode Function

scholarly journals A Deep Neural Network for Unsupervised Anomaly Detection and Diagnosis in Multivariate Time Series Data

2019 ◽  
Vol 33 ◽  
pp. 1409-1416 ◽  
Author(s):  
Chuxu Zhang ◽  
Dongjin Song ◽  
Yuncong Chen ◽  
Xinyang Feng ◽  
Cristian Lumezanu ◽  
...  
Keyword(s):  
Time Series ◽  
Anomaly Detection ◽  
Power Plants ◽  
Time Series Data ◽  
Multivariate Time Series ◽  
Series Data ◽  
Feature Maps ◽  
Multi Scale ◽  
Unsupervised Anomaly Detection ◽  
Detection And Diagnosis

Nowadays, multivariate time series data are increasingly collected in various real world systems, e.g., power plants, wearable devices, etc. Anomaly detection and diagnosis in multivariate time series refer to identifying abnormal status in certain time steps and pinpointing the root causes. Building such a system, however, is challenging since it not only requires to capture the temporal dependency in each time series, but also need encode the inter-correlations between different pairs of time series. In addition, the system should be robust to noise and provide operators with different levels of anomaly scores based upon the severity of different incidents. Despite the fact that a number of unsupervised anomaly detection algorithms have been developed, few of them can jointly address these challenges. In this paper, we propose a Multi-Scale Convolutional Recurrent Encoder-Decoder (MSCRED), to perform anomaly detection and diagnosis in multivariate time series data. Specifically, MSCRED first constructs multi-scale (resolution) signature matrices to characterize multiple levels of the system statuses in different time steps. Subsequently, given the signature matrices, a convolutional encoder is employed to encode the inter-sensor (time series) correlations and an attention based Convolutional Long-Short Term Memory (ConvLSTM) network is developed to capture the temporal patterns. Finally, based upon the feature maps which encode the inter-sensor correlations and temporal information, a convolutional decoder is used to reconstruct the input signature matrices and the residual signature matrices are further utilized to detect and diagnose anomalies. Extensive empirical studies based on a synthetic dataset and a real power plant dataset demonstrate that MSCRED can outperform state-ofthe-art baseline methods.

Sign in / Sign up

Export Citation Format

Share Document