Dynamic Segmentation Of Youth Accelerometer Data By Sojourn And Change Point Detection Methods

Introduction: Data from wrist-worn accelerometers often has an inherent natural segmentation that reflects transitioning from one activity to another. The aim of this study was to develop an activity transition detection method to realize this natural segmentation. Methods: Data was gathered from 16 participants who wore triaxial wrist accelerometers in a lab-based protocol and 47 participants in a free-living protocol. Change point detection was used to create a method for detecting activity transitions. The agreement between observed and predicted transitions was assessed by the Matthews Correlation Coefficient (MCC), Root Mean Squared Error (RMSE), and two additional metrics created for this task; the Ratio of Minimum Mean Distance (RMMD) and the Ratio of Sensitivity (RoS). The effects of varying combinations of acceleration axes were also investigated to determine the most effective set of axes. A novel post-processing technique was developed to mitigate a major limitation identified in current transition detection methods. Results: The developed transition detection method achieved a MCC of 0.763, a RMSE of 3.17, a RoS of 2.40, and a RMMD of 3.21, outperforming existing techniques. The post-processing technique developed improved the performance of all methods when identifying transitions. It was found that using solely the y-axis (vertical acceleration) allowed for optimal performance. Conclusion: Change point detection is a valid method for identifying transitions in activity using wrist-worn accelerometer data. The new post processing technique developed improves the performance of transition detection methods.

Download Full-text

Sequential Change-Point Detection Methods for Nonstationary Time Series

Technometrics ◽

10.1198/004017007000000434 ◽

2008 ◽

Vol 50 (1) ◽

pp. 40-52 ◽

Cited By ~ 14

Author(s):

Hyunyoung Choi ◽

Hernando Ombao ◽

Bonnie Ray

Keyword(s):

Time Series ◽

Change Point ◽

Nonstationary Time Series ◽

Change Point Detection ◽

Detection Methods ◽

Sequential Change ◽

Point Detection

Download Full-text

Comparison between change point detection methods with synthetic rainfall data and application in South Korea

KSCE Journal of Civil Engineering ◽

10.1007/s12205-014-0228-5 ◽

2014 ◽

Vol 20 (4) ◽

pp. 1558-1571 ◽

Cited By ~ 6

Author(s):

Sangho Lee ◽

Sang Ug Kim

Keyword(s):

South Korea ◽

Change Point ◽

Change Point Detection ◽

Rainfall Data ◽

Detection Methods ◽

Point Detection

Download Full-text

Change-point detection in wind turbine SCADA data for robust condition monitoring with normal behaviour models

10.5194/wes-2020-38 ◽

2020 ◽

Author(s):

Simon Letzgus

Keyword(s):

Wind Turbine ◽

Change Point ◽

Significant Loss ◽

Change Point Detection ◽

Training Data ◽

Change Points ◽

Detection Methods ◽

Full Potential ◽

Normal Behaviour ◽

Point Detection

Abstract. Analysis of data from wind turbine supervisory control and data acquisition (SCADA) systems has attracted considerable research interest in recent years. The data is predominantly used to gain insights into turbine condition without the need for additional sensing equipment. Most successful approaches apply semi-supervised anomaly detection methods, also called normal behaivour models, that use clean training data sets to establish healthy component baseline models. However, one of the major challenges when working with wind turbine SCADA data in practice is the presence of systematic changes in signal behaviour induced by malfunctions or maintenance actions. Even though this problem is well described in literature it has not been systematically addressed so far. This contribution is the first to comprehensively analyse the presence of change-points in wind turbine SCADA signals and introduce an algorithm for their automated detection. 600 signals from 33 turbines are analysed over an operational period of more than two years. During this time one third of the signals are affected by change-points. Kernel change-point detection methods have shown promising results in similar settings but their performance strongly depends on the choice of several hyperparameters. This contribution presents a comprehensive comparison between different kernels as well as kernel-bandwidth and regularisation-penalty selection heuristics. Moreover, an appropriate data pre-processing procedure is introduced. The results show that the combination of Laplace kernels with a newly introduced bandwidth and penalty selection heuristic robustly outperforms existing methods. In a signal validation setting more than 90 % of the signals were classified correctly regarding the presence or absence of change-points, resulting in a F1-score of 0.86. For a change-point-free sequence selection the most severe 60 % of all CPs could be automatically removed with a precision of more than 0.96 and therefore without a significant loss of training data. These results indicate that the algorithm can be a meaningful step towards automated SCADA data pre-processing which is key for data driven methods to reach their full potential. The algorithm is open source and its implementation in Python publicly available.

Download Full-text

THE REVIEW OF CHANGE-POINT DETECTION METHODS AND PROSPECTS OF THEIR APPLICATION IN THE TECHNICAL ANALYSIS OF STOCK EXCHANGE QUOTES

Actual problems of automation and information technology ◽

10.15421/431210 ◽

2011 ◽

Vol 16 (0) ◽

Author(s):

O. P. Lutsenko ◽

O. H. Baybuz

Keyword(s):

Change Point ◽

Stock Exchange ◽

Technical Analysis ◽

Change Point Detection ◽

Detection Methods ◽

Point Detection

Download Full-text

Machine Committee Framework for Power Grid Disturbances Analysis Using Synchrophasors Data

Smart Cities ◽

10.3390/smartcities4010001 ◽

2020 ◽

Vol 4 (1) ◽

pp. 1-16

Author(s):

Haoran Niu ◽

Olufemi A. Omitaomu ◽

Qing C. Cao

Keyword(s):

Event Detection ◽

Change Point ◽

Power Grid ◽

Detection Algorithm ◽

Change Point Detection ◽

Detection Methods ◽

Detection Accuracy ◽

Disturbance Analysis ◽

Events Detection ◽

Point Detection

Events detection is a key challenge in power grid frequency disturbances analysis. Accurate detection of events is crucial for situational awareness of the power system. In this paper, we study the problem of events detection in power grid frequency disturbance analysis using synchrophasors data streams. Current events detection approaches for power grid rely on individual detection algorithm. This study integrates some of the existing detection algorithms using the concept of machine committee to develop improved detection approaches for grid disturbance analysis. Specifically, we propose two algorithms—an Event Detection Machine Committee (EDMC) algorithm and a Change-Point Detection Machine Committee (CPDMC) algorithm. Both algorithms use parallel architecture to fuse detection knowledge of its individual methods to arrive at an overall output. The EDMC algorithm combines five individual event detection methods, while the CPDMC algorithm combines two change-point detection methods. Each method performs the detection task separately. The overall output of each algorithm is then computed using a voting strategy. The proposed algorithms are evaluated using three case studies of actual power grid disturbances. Compared with the individual results of the various detection methods, we found that the EDMC algorithm is a better fit for analyzing synchrophasors data; it improves the detection accuracy; and it is suitable for practical scenarios.

Download Full-text

Application of multiple change point detection methods to large urban telecommunication networks

2017 28th Irish Signals and Systems Conference (ISSC) ◽

10.1109/issc.2017.7983608 ◽

2017 ◽

Cited By ~ 1

Author(s):

Andrew Shields ◽

Pat Doody ◽

Ted Scully

Keyword(s):

Change Point ◽

Telecommunication Networks ◽

Change Point Detection ◽

Detection Methods ◽

Point Detection

Download Full-text

DEVELOPING TIME-BASED CLUSTERING NEURAL NETWORKS TO USE CHANGE-POINT DETECTION: APPLICATION TO FINANCIAL TIME SERIES

Asia Pacific Journal of Operational Research ◽

10.1142/s0217595905000431 ◽

2005 ◽

Vol 22 (01) ◽

pp. 51-70 ◽

Cited By ~ 8

Author(s):

KYONG JOO OH ◽

TAE HYUP ROH ◽

MYUNG SANG MOON

Keyword(s):

Neural Network ◽

Neural Networks ◽

Time Series ◽

Change Point ◽

Financial Time Series ◽

Change Point Detection ◽

Detection Methods ◽

Financial Time ◽

Point Groups ◽

Point Detection

This study suggests time-based clustering models integrating change-point detection and neural networks, and applies them to financial time series forecasting. The basic concept of the proposed models is to obtain intervals divided by change points, to identify them as change-point groups, and to involve them in the forecasting model. The proposed models consist of two stages. The first stage, the clustering neural network modeling stage, is to detect successive change points in the dataset, and to forecast change-point groups with backpropagation neural networks (BPNs). In this stage, three change-point detection methods are applied and compared. They are: (1) the parametric approach, (2) the nonparametric approach, and (3) the model-based approach. The next stage is to forecast the final output with BPNs. Through the application to financial time series forecasting, we compare the proposed models with a neural network model alone and, in addition, determine which of three change-point detection methods performs better. Furthermore, we evaluate whether the proposed models play a role in clustering to reflect the time. Finally, this study examines the predictability of the integrated neural network models based on change-point detection.

Download Full-text