Optimization of Time-Series Data Partitioning for Parameter Identification

2010 ◽  
Author(s):  
Soumik Sarkar ◽  
Kushal Mukherjee ◽  
Xin Jin ◽  
Asok Ray
Keyword(s):  
Time Series ◽  
Parameter Identification ◽  
Time Series Data ◽  
Nonlinear Dynamical System ◽  
Data Partitioning ◽  
Set Partitioning ◽  
Series Data ◽  
Data Set ◽  
Partitioning Procedure ◽  
Symbol Sequences

This paper presents a data-driven method of parameter identification in nonlinear systems based on the theories of symbolic dynamics. Although construction of finite-state-machine models from symbol sequences has been widely reported, similar efforts have not been expended to investigate partitioning of time series data to optimally generate symbol sequences. A data-set partitioning procedure is proposed to extract features from time series data by optimizing a multi-objective cost functional. Performance of the optimal partitioning procedure is compared with those of other traditional partitioning (e.g., uniform and maximum entropy) schemes. Then, tools of pattern classification are applied to identify the ranges of multiple parameters of a well-known chaotic nonlinear dynamical system, namely the Duffing Equation, from its time series response.

scholarly journals Remaining Useful Life Prediction Using Temporal Convolution with Attention

AI ◽  
2021 ◽  
Vol 2 (1) ◽  
pp. 48-70
Author(s):  
Wei Ming Tan ◽  
T. Hui Teo
Keyword(s):  
Neural Network ◽  
Time Series ◽  
Time Series Data ◽  
Remaining Useful Life ◽  
Sensor Data ◽  
Series Data ◽  
Multiple Time ◽  
Data Set ◽  
Form Complex ◽  
Useful Life

Prognostic techniques attempt to predict the Remaining Useful Life (RUL) of a subsystem or a component. Such techniques often use sensor data which are periodically measured and recorded into a time series data set. Such multivariate data sets form complex and non-linear inter-dependencies through recorded time steps and between sensors. Many current existing algorithms for prognostic purposes starts to explore Deep Neural Network (DNN) and its effectiveness in the field. Although Deep Learning (DL) techniques outperform the traditional prognostic algorithms, the networks are generally complex to deploy or train. This paper proposes a Multi-variable Time Series (MTS) focused approach to prognostics that implements a lightweight Convolutional Neural Network (CNN) with attention mechanism. The convolution filters work to extract the abstract temporal patterns from the multiple time series, while the attention mechanisms review the information across the time axis and select the relevant information. The results suggest that the proposed method not only produces a superior accuracy of RUL estimation but it also trains many folds faster than the reported works. The superiority of deploying the network is also demonstrated on a lightweight hardware platform by not just being much compact, but also more efficient for the resource restricted environment.

scholarly journals Studying monthly rainfall over Dibrugarh, Assam: Use of SARIMA approach

MAUSAM ◽  
2021 ◽  
Vol 68 (2) ◽  
pp. 349-356
Author(s):  
J. HAZARIKA ◽  
B. PATHAK ◽  
A. N. PATOWARY
Keyword(s):  
Time Series ◽  
Time Series Data ◽  
Moving Average ◽  
Demand Management ◽  
Arima Model ◽  
Monthly Rainfall ◽  
Series Data ◽  
Data Set ◽  
Modeling And Forecasting ◽  
Moving Average Model

Perceptive the rainfall pattern is tough for the solution of several regional environmental issues of water resources management, with implications for agriculture, climate change, and natural calamity such as floods and droughts. Statistical computing, modeling and forecasting data are key instruments for studying these patterns. The study of time series analysis and forecasting has become a major tool in different applications in hydrology and environmental fields. Among the most effective approaches for analyzing time series data is the ARIMA (Autoregressive Integrated Moving Average) model introduced by Box and Jenkins. In this study, an attempt has been made to use Box-Jenkins methodology to build ARIMA model for monthly rainfall data taken from Dibrugarh for the period of 1980- 2014 with a total of 420 points.  We investigated and found that ARIMA (0, 0, 0) (0, 1, 1)12 model is suitable for the given data set. As such this model can be used to forecast the pattern of monthly rainfall for the upcoming years, which can help the decision makers to establish priorities in terms of agricultural, flood, water demand management etc.  

Use of Temporal Irreversibility of Symbolic Time Series for Early Detection of Extinction in Thermal Pulse Combustors

2006 ◽  
Author(s):  
Subhashis Datta ◽  
Achintya Mukhopadhyay ◽  
Dipankar Sanyal
Keyword(s):  
Time Series ◽  
Fourth Order ◽  
Time Series Data ◽  
Order System ◽  
Series Data ◽  
Thermal Pulse ◽  
Pulse Combustor ◽  
The Difference ◽  
Symbol Sequences ◽  
System Approaches

A nonlinear fourth-order dynamic model of a thermal pulse combustor has been developed. In this work, the time series data generated by solution of the fourth order system is converted into a set of symbols based on the values of pressure variables. The key step to symbolization involves transformation of the original values to a stream of discretised symbols by partitioning the range of observed values into a finite number of regions and then assigning a symbol to each measurement based on the region in which it falls. Once all the measured values are symbolized, a symbol sequence vector consisting of L successive temporal observations is defined and its relative frequency is determined. In this work, the relative frequencies of different symbol sequences are computed by scanning the time series data in forward and reverse directions. The difference between the relative frequencies obtained in forward and reverse scanning is termed as "irreversibility" of the process. It is observed that for given alphabet and word sizes, the "irreversibility" increases as the system approaches extinction. The effects of different choices of alphabet and word sizes are also considered.

Exploratory Time Series Data Mining by Genetic Clustering

2011 ◽  
pp. 157-178
Author(s):  
T. Warren Liao
Keyword(s):  
Data Mining ◽  
Time Series ◽  
Time Series Data ◽  
Distance Measures ◽  
Series Data ◽  
Synthetic Control ◽  
Data Set ◽  
Univariate Time Series ◽  
Genetic Clustering ◽  
Data Objects

In this chapter, we present genetic algorithm (GA) based methods developed for clustering univariate time series with equal or unequal length as an exploratory step of data mining. These methods basically implement the k-medoids algorithm. Each chromosome encodes in binary the data objects serving as the k-medoids. To compare their performance, both fixed-parameter and adaptive GAs were used. We first employed the synthetic control chart data set to investigate the performance of three fitness functions, two distance measures, and other GA parameters such as population size, crossover rate, and mutation rate. Two more sets of time series with or without known number of clusters were also experimented: one is the cylinder-bell-funnel data and the other is the novel battle simulation data. The clustering results are presented and discussed.

A simple method for reconstructing a high-quality NDVI time-series data set based on the Savitzky–Golay filter

2004 ◽  
Vol 91 (3-4) ◽  
pp. 332-344 ◽  
Author(s):  
Jin Chen ◽  
Per. Jönsson ◽  
Masayuki Tamura ◽  
Zhihui Gu ◽  
Bunkei Matsushita ◽  
...  
Keyword(s):  
Time Series ◽  
Time Series Data ◽  
Series Data ◽  
High Quality ◽  
Data Set ◽  
Simple Method ◽  
Ndvi Time Series

Condition Monitoring of Equipment in Oil Wells using Deep Learning

2020 ◽  
Vol 12 (01) ◽  
pp. 2050001
Author(s):  
Yadigar N. Imamverdiyev ◽  
Fargana J. Abdullayeva
Keyword(s):  
Time Series ◽  
Deep Learning ◽  
Time Series Data ◽  
Short Term Memory ◽  
Single Layer ◽  
Oil Wells ◽  
Fault Prediction ◽  
Series Data ◽  
Oil Well ◽  
Data Set

In this paper, a fault prediction method for oil well equipment based on the analysis of time series data obtained from multiple sensors is proposed. The proposed method is based on deep learning (DL). For this purpose, comparative analysis of single-layer long short-term memory (LSTM) with the convolutional neural network (CNN) and stacked LSTM methods is provided. To demonstrate the efficacy of the proposed method, some experiments are conducted on the real data set obtained from eight sensors installed in oil wells. In this paper, compared to the single-layer LSTM model, the CNN and stacked LSTM predicted the faulty time series with a minimal loss.

Rough Sets Based Prediction Model of Tick-Wise Price Fluctuations

2011 ◽  
Vol 15 (4) ◽  
pp. 449-453 ◽  
Author(s):  
Yoshiyuki Matsumoto ◽  
◽  
Junzo Watada ◽  
Keyword(s):  
Time Series ◽  
Decision Rule ◽  
Rough Set ◽  
Rough Sets ◽  
Time Series Data ◽  
Series Data ◽  
Data Set ◽  
Price Fluctuations ◽  
Rough Sets Theory ◽  
Sets Theory

Rough sets theory was proposed by Z. Pawlak in 1982. This theory enables us to mine knowledge granules through a decision rule from a database, a web base, a set and so on. We can apply the decision rule to reason, estimate, evaluate, or forecast unknown objects. In this paper, the rough set model is used to analyze of time series data of tick-wise price fluctuation, where knowledge granules are mined from the data set of tick-wise price fluctuations.

Sign in / Sign up

Export Citation Format

Share Document