Multi-Dimensional Encoding Image Based Time Series Feature Extraction for Improved Colorimetric Gas Concentration Estimation

Purpose In this study, a physical similarity simulation plays a significant role in the study of crack evolution and the gas migration mechanism. A sensor is deployed inside a comparable artificial rock formation to assure the accuracy of the experiment results. During the building of the simulated rock formation, a huge volume of acidic gas is released, causing numerous sensor measurement mistakes. Additionally, the gas concentration estimation approach is subject to uncertainty because of the complex rock formation environment. As a result, the purpose of this study is to introduce an adaptive Kalman filter approach to reduce observation noise, increase the accuracy of the gas concentration estimation model and, finally, determine the gas migration law. Design/methodology/approach First, based on the process of gas floatation-diffusion and seepage, the gas migration model is established according to Fick’s second law, and a simplified modeling method using diffusion flux instead of gas concentration is presented. Second, an adaptive Kalman filter algorithm is introduced to establish a gas concentration estimation model, taking into account the model uncertainty and the unknown measurement noise. Finally, according to a large-scale physical similarity simulation platform, a thorough experiment about gas migration is carried out to extract gas concentration variation data with certain ventilation techniques and to create a gas chart of the time-changing trend. Findings This approach is used to determine the changing process of gas distribution for a certain ventilation mode. The results match the rock fissure distribution condition derived from the microseismic monitoring data, proving the effectiveness of the approach. Originality/value For the first time in large-scale three-dimensional physical similarity simulations, the adaptive Kalman filter data processing method based on the inverse Wishart probability density function is used to solve the problem of an inaccurate process and measurement noise, laying the groundwork for studying the gas migration law and determining the gas migration mechanism.

Download Full-text

Extracting Features from Time Series

Fundamentals of Clinical Data Science ◽

10.1007/978-3-319-99713-1_7 ◽

2018 ◽

pp. 85-100 ◽

Cited By ~ 1

Author(s):

Christian Herff ◽

Dean J. Krusienski

Keyword(s):

Time Series ◽

Feature Extraction ◽

Clinical Data ◽

Noise Filtering ◽

Time Intervals ◽

Time Points ◽

Biomedical Systems ◽

Patient Weight ◽

Short Time

AbstractClinical data is often collected and processed as time series: a sequence of data indexed by successive time points. Such time series can be from sources that are sampled over short time intervals to represent continuous biophysical wave-(one word waveforms) forms such as the voltage measurements representing the electrocardiogram, to measurements that are sampled daily, weekly, yearly, etc. such as patient weight, blood triglyceride levels, etc. When analyzing clinical data or designing biomedical systems for measurements, interventions, or diagnostic aids, it is important to represent the information contained within such time series in a more compact or meaningful form (e.g., noise filtering), amenable to interpretation by a human or computer. This process is known as feature extraction. This chapter will discuss some fundamental techniques for extracting features from time series representing general forms of clinical data.

Download Full-text

Anomaly Detection for Time Series with Difference Rate Sample Entropy and Generative Adversarial Networks

Complexity ◽

10.1155/2021/5854096 ◽

2021 ◽

Vol 2021 ◽

pp. 1-13

Author(s):

Keke Gao ◽

Wenbin Feng ◽

Xia Zhao ◽

Chongchong Yu ◽

Weijun Su ◽

...

Keyword(s):

Time Series ◽

Anomaly Detection ◽

Normal Mode ◽

Coal Mine ◽

Spontaneous Combustion ◽

Sample Entropy ◽

Generative Adversarial Networks ◽

Detection Accuracy ◽

Gas Concentration ◽

Adversarial Networks

The spontaneous combustion of residual coals in the mined-out area tends to cause an explosion, which is one kind of severe thermodynamic compound disaster of coal mines and leads to serious losses to people's lives and production safety. The prediction and early warning of coal mine thermodynamic disasters are mainly determined by the changes of the index gas concentration pattern in coal mine mined-out areas collected continuously. The time series anomaly pattern detection method is mainly used to reach the state change of gas concentration pattern. The change of gas concentration follows a certain rule as time changes. A great change in the gas concentration indicates the possibility of coal spontaneous combustion and other disasters. To emphasize the features of collected maker gas and overcome the low anomaly detection accuracy caused by the inadequate learning of the normal mode, this paper adopted a method of anomaly detection for time series with difference rate sample entropy and generative adversarial networks. Because the difference rate entropy feature of abnormal data was much larger than that of normal mode, this paper improved the calculation method of the abnormal score by giving different weights to the detection points to enhance the detection rate. To verify the effectiveness of the proposed method, this paper employed simulation models of the mined-out area and adopted coal samples from Dafosi Coal Mine to carry out experiments. Preliminary testing was performed using monitoring data from a coal mine. The experiment compared the entropy results of different time series with the detection results of generative adversarial networks and automatic encoders and showed that the method proposed in this paper had relatively high detection accuracy.

Download Full-text

Feature extraction from time series data

Journal of Computational Methods in Sciences and Engineering ◽

10.3233/jcm-2009-0239 ◽

2009 ◽

Vol 9 (s1) ◽

pp. S99-S110 ◽

Cited By ~ 2

Author(s):

Durga Toshniwal

Keyword(s):

Time Series ◽

Feature Extraction ◽

Time Series Data ◽

Series Data

Download Full-text

Integrating Data-Driven Segmentation, Local Feature Extraction and Fisher Kernel Encoding to Improve Time Series Classification

Neural Processing Letters ◽

10.1007/s11063-018-9798-4 ◽

2018 ◽

Vol 49 (1) ◽

pp. 43-66

Author(s):

Weiping Huang ◽

Boxuan Yue ◽

Qinghua Chi ◽

Jun Liang

Keyword(s):

Time Series ◽

Feature Extraction ◽

Local Feature ◽

Data Driven ◽

Time Series Classification ◽

Fisher Kernel ◽

Local Feature Extraction

Download Full-text

A Linear Time Series Analysis of Fetal Heart Rate to Detect the Variability

Handbook of Research on Recent Developments in Intelligent Communication Application - Advances in Wireless Technologies and Telecommunication ◽

10.4018/978-1-5225-1785-6.ch018 ◽

2017 ◽

pp. 471-495 ◽

Cited By ~ 4

Author(s):

Sahana Das ◽

Kaushik Roy ◽

Chanchal Kumar Saha

Keyword(s):

Heart Rate ◽

Time Series ◽

Feature Extraction ◽

Time Series Analysis ◽

Real Time ◽

Fetal Heart Rate ◽

Fetal Heart ◽

Linear Time ◽

Automated Diagnosis ◽

Real Time Analysis

Real time analysis and interpretation of fetal heart rate (FHR) is the challenge posed to every clinician. Different algorithms had been developed, tried and subsequently incorporated into Cardiotocograph (CTG) machines for automated diagnosis. Feature extraction and accurate detection of baseline and its variability has been the focus of this chapter. Algorithms by Dawes and Redman and Ayres-de-Campos have been discussed in this chapter. The authors are pleased to propose an algorithm for extracting the variability of fetal heart. The algorithm's accuracy and degree of agreement with clinician's diagnosis had been established by various statistical methods. This algorithm has been compared with an algorithm proposed by Nidhal and the new algorithm is found to be better at detecting variability in both ante-partum and intra-partum period.

Download Full-text