Research on a Method of Fault Signal Extraction Based on Improved Algorithm

2012 ◽  
Vol 217-219 ◽  
pp. 2692-2696
Author(s):  
Ying Wang ◽  
You Rong Li ◽  
Xiao Qin Zhu ◽  
Pan Lin ◽  
Yue Sheng Luo
Keyword(s):  
Time Series ◽  
Feature Extraction ◽  
Partial Information ◽  
Initial Time ◽  
Time Signal ◽  
Shock Attenuation ◽  
Information Time ◽  
Initial Time Series ◽  
Fault Information ◽  
Sequential Decomposition

Considering the difficulty of diagnosis signal de-noising and feature extraction problems, according to the characteristics of periodicity and shock attenuation respond of mechanical fault vibration signals, a method of improved sequential decomposition algorithm is proposed, it transforms an initial time series into a group of two-dimensional time series, prominent time series partial information, time series decomposition is reversible, can be used for filtering and feature extraction of time signal. Through the simulation and experiments, the validity of method for highlighting partial feature information of the signal is verified, helping to extract weak fault information in strong background noise environment.

scholarly journals A forecasting the consumer price index using time series model

2021 ◽  
Vol 107 ◽  
pp. 10002
Author(s):  
Volodymyr Shinkarenko ◽  
Alexey Hostryk ◽  
Larysa Shynkarenko ◽  
Leonid Dolinskyi
Keyword(s):  
Time Series ◽  
Price Index ◽  
Consumer Price Index ◽  
Exponential Smoothing ◽  
Time Series Model ◽  
Initial Time ◽  
Autocorrelation Functions ◽  
Minimum Error ◽  
Initial Time Series

This article examines the behavior of the consumer price index in Ukraine for the period from January 2010 to September 2020. The characteristics of the initial time series, the analysis of autocorrelation functions made it possible to reveal the tendency of their development and the presence of annual seasonality. To model the behavior of the consumer price index and forecast for the next months, two types of models were used: the additive ARIMA*ARIMAS model, better known as the model of Box-Jenkins and the exponential smoothing model with the seasonality estimate of Holt-Winters. As a result of using the STATISTICA package, the most adequate models were built, reflecting the monthly dynamics of the consumer price index in Ukraine. The inflation forecast was carried out on the basis of the Holt-Winters model, which has a minimum error.

scholarly journals Forecasting The Exchange Rate Between Euro And USD: Probabilistic Approach Versus ARIMA And Exponential Smoothing Techniques

2012 ◽  
Vol 28 (2) ◽  
pp. 171 ◽  
Author(s):  
Paraschos Maniatis
Keyword(s):  
Time Series ◽  
Exchange Rate ◽  
Probabilistic Approach ◽  
Laplace Distribution ◽  
Exponential Smoothing ◽  
Initial Time ◽  
Font Size ◽  
Smoothing Techniques ◽  
Initial Time Series ◽  
The Exchange Rate

<span style="font-family: Times New Roman; font-size: small;"> </span><p style="margin: 0in 0.5in 0pt; text-align: justify; mso-pagination: none;" class="MsoNoSpacing"><span style="color: black; font-family: &quot;Times New Roman&quot;,&quot;serif&quot;; font-size: 10pt; mso-themecolor: text1; mso-ansi-language: EN-US;">This study attempts to model the exchange rate between Euro and USD using univariate models- in particular ARIMA and exponential smoothing techniques. The time series analysis reveals non stationarity in data and, therefore, the models fail to give reliable predictions. However, differencing the initial time series the resulting series shows strong resemblance to white noise. The analysis of this series advocates independence in data and distribution satisfactorily close to Laplace distribution. The application of Laplace distribution offers reliable probabilities in forecasting changes in the exchange rate.</span></p><span style="font-family: Times New Roman; font-size: small;"> </span>

scholarly journals Generalized Relational Tensors For Irregularly Sampled Time Series: Methods To Store, Re-generate, And Predict

2021 ◽  
Author(s):  
Vasilii Gromov ◽  
Anastasia Necheporenko ◽  
Andrei Gaisin ◽  
Ilya Volkov ◽  
Stanislav Diner
Keyword(s):  
Time Series ◽  
Multivariate Time Series ◽  
Chaotic Time Series ◽  
Optimization Techniques ◽  
Initial Time ◽  
Auto Correlation ◽  
Initial Time Series ◽  
Auto Correlation Function ◽  
Estimate Quality

Abstract The paper deals with a generalized relational tensor, a novel discrete structure to store information about a time series, and algorithms (1) to fill the structure, (2) to generate a time series from the structure, and (3) to predict a time series, for both regularly and irregularly sampled time series. The algorithms combine the concept of generalized z-vectors with ant colony optimization techniques. In order to estimate quality of the storing/re-generating procedure, a difference between characteristics of the initial and regenerated time series is used. The structure allows working with a multivariate time series, with an irregularly sampled time series, and with a number of series as well. For chaotic time series, a difference between characteristics of the initial time series (the highest Lyapunov exponent, the auto-correlation function) and those of the time series re-generated from a structure is used to assess the effectiveness of the algorithms in question. The approach has shown fairly good results for periodic and benchmark chaotic time series and satisfactory results for real-world chaotic data.

scholarly journals FORECASTING THE GROWTH RATE OF GULLIES IN THE UDMURT REPUBLIC

2019 ◽  
Vol 29 (3) ◽  
pp. 383-393
Author(s):  
M.Yu. Zaitseva ◽  
I.I. Rysin
Keyword(s):  
Growth Rate ◽  
Time Series ◽  
Linear Growth ◽  
Extreme Values ◽  
Initial Time ◽  
Linear Growth Rate ◽  
Medium Term ◽  
Term Forecast ◽  
Initial Time Series ◽  
Udmurt Republic

The present study is aimed at forecasting the processes of gullyerosion in the Udmurt Republic using the methods of mathematical modeling. Five time series characterizing the average linear growth rate of gullies for the period from 1978 to 2017 were selected as a source material. Gullies were grouped according to the geographical principle and genesis. As part of this work, it is expected to build a medium-term forecast for the period 2018-2022. Fourier analysis was chosen as the basis for working with the initial time series. The results of the obtained models are graphically displayed. Subsequent regression analysis confirmed the validity of the model for at least four of the five groups of gullies. However, when comparing the obtained forecast values with those actually measured in 2018, it turned out that this model could not take into account the possible extreme values of the growth of individual gullies in the group.

scholarly journals Anomaly Detection and Classification of Physiological Signals in IoT- Based Healthcare Framework

2021 ◽  
Author(s):  
Menaa Nawaz ◽  
Jameel Ahmed
Keyword(s):  
Time Series ◽  
Feature Extraction ◽  
Deep Learning ◽  
Anomaly Detection ◽  
Real Time ◽  
Signal Analysis ◽  
Physiological Signals ◽  
Series Data ◽  
Time Signal ◽  
Biomedical Sensors

Abstract Physiological signals retrieve information from sensors implanted or attached to the human body. These signals are vital data sources that can assist in predicting the disease well before time; thus, proper treatment can be made possible. With the addition of the Internet of Things in healthcare, real-time data collection and preprocessing for signal analysis has reduced the burden of in-person appointments and decision making on healthcare. Recently, deep learning-based algorithms have been implemented by researchers for the recognition, realization and prediction of diseases by extracting and analyzing important features. In this research, real-time 1-D time series data of on-body noninvasive biomedical sensors were acquired, preprocessed and analysed for anomaly detection. Feature engineered parameters of large and diverse datasets have been used to train the data to make the anomaly detection system more reliable. For comprehensive real-time monitoring, the implemented system uses wavelet time scattering features for classification and a deep learning-based autoencoder for anomaly detection of time series signals to assist the clinical diagnosis of cardiovascular and muscular activity. In this research, an implementation of an IoT-based AI-edge healthcare framework using biomedical sensors was presented. This paper also aims to analyse cloud data acquired through biomedical sensors using signal analysis techniques for anomaly detection, and time series classification has been performed for disease prognosis in real time by implementing 24 AI-based techniques to find the most accurate technique for real-time raw signals. The deep learning-based LSTM method based on wavelet time scattering feature extraction has shown a classification test accuracy of 100%. Using wavelet time scattering feature extraction achieved 95% signal reduction to increase the real-time processing speed. In real-time signal anomaly detection, 98% accuracy is achieved using LSTM autoencoders. The average mean absolute error loss of 0.0072 for normal signals and 0.078 is achieved for anomalous signals.

scholarly journals EUR/RON EXCHANGE RATE PREDICTION BASED ON BOX-JENKINS TECHNIQUE

2019 ◽  
Vol 1 (2) ◽  
pp. 31-41
Author(s):  
Costică Ionela ◽  
Boitan Iustina Alina
Keyword(s):  
Time Series ◽  
Exchange Rate ◽  
Goodness Of Fit ◽  
Moving Average ◽  
Model Specification ◽  
Initial Time ◽  
Main Research ◽  
Rate Analysis ◽  
Initial Time Series ◽  
The Exchange Rate

The aim of this study consists in analyzing the importance of the exchange rate forecast using the Box-Jenkins models, also known as Auto Regressive Integrated Moving Average (ARIMA) models. The first part of the paper presents the main research in this field, which can be classified in two categories (studies applying classical methods, such as Box-Jenkins models and studies which rely on sophisticated prediction tools), and summarizes the main findings of some of the studies applying Box-Jenkins models. In the second part of the paper we performed a EUR/RON exchange rate analysis and forecasting, based on testing several AR, MA and ARMA candidate processes, in order to find out the best fitting model specification.  We adopted the following strategy: i) an initial time series had been used for testing various model specifications, identify the best performing one and making a forecast of the EUR/RON exchange rate; ii) after comparing the accuracy of this forecast with the real level recorded by the exchange rate at end of May 2018, we conducted a second forecast, for the period May 2019 – November 2019. The initial time series employed has daily frequency and covers the timeframe July 4, 2005 – December 5, 2017, while the second time series used covers the period July 4, 2005 – May 6, 2019. The empirical findings have passed the goodness-of-fit tests and show a good predictive power. The first forecast performed for a six month period (December 2017 – May 2018) has indicated a slow pace, persistent increase of the EUR/RON exchange rate, which was confirmed by the expectations of market participants (financial analysts, banks’ research departments). The second forecast, which covers the period May 2019 – November 2019, indicates a similar rising trend and the ongoing depreciation of the national currency.

Sign in / Sign up

Export Citation Format

Share Document