scholarly journals The application of phase difference to analysis the magnetotelluric data

2021 ◽  
Author(s):  
Hao Chen ◽  
Hideki Mizunaga ◽  
Toshiaki Tanaka ◽  
Lei Zhou
Keyword(s):  
Frequency Domain ◽  
Phase Difference ◽  
Series Data ◽  
Selection Strategy ◽  
Impedance Tensor ◽  
Field Variation ◽  
Geoelectric Field ◽  
Polarization Direction ◽  
Contaminated Data ◽  
M Estimator

Abstract Magnetotelluric (MT) method is an electromagnetic geophysical method for inferring the earth's subsurface electrical conductivity from measurements of natural geomagnetic and geoelectric field variation at the earth's surface. The first step in MT data processing is to estimate the impedance tensor in the frequency domain from the measured time-series data. The initial MT response function estimator is based on the least-square theory; it can be severely disturbed by the cultural noise. In the presence of a small amount of intermittent contaminated data, it can be improved by remote reference technique, robust procedure or combination of them. In the presence of a large amount of contaminated data, it can still succeed with assistance from data analysis to remove the most contaminated data before the impedance tensor estimation. The phase difference is an important parameter to analyze the data in the frequency domain. In this paper, we investigate three parameters(the predicted coherence, remote coherence and polarization direction) correspond the phase difference to analyze the MT data. We demonstrated that the high predicted coherence could indicate a high signal-to-noise ratio(SNR) or strong coherence noise. The polarization direction was useful to visualize the background noise. The remote coherence was a useful parameter to indicate the quality of the data. In this paper, we will introduce a robust M-estimator at first. At last, we showed the effectiveness of the application of remote linear coherence to the selection strategy based on the M-estimator. By this selection strategy, the result can be improved dramatically in the presence of a large amount of intermittent noise.

scholarly journals New Strategy to Calculate Robust Impedance using RMHHT Estimator for Continuous and Discontinuous Broadband Magnetotelluric Time-Series Data

2020 ◽  
Author(s):  
Hao Chen ◽  
Hideki Mizunaga ◽  
Toshiaki Tanaka ◽  
Gang Wang
Keyword(s):  
Time Series ◽  
Urban Areas ◽  
Time Series Data ◽  
Computation Time ◽  
Series Data ◽  
Field Variation ◽  
Geoelectric Field ◽  
Quiet Time ◽  
New Strategy ◽  
Electric Equipment

Abstract Magnetotelluric (MT) method is an electromagnetic geophysical method for inferring the earth's subsurface electrical conductivity from measurements of natural geomagnetic and geoelectric field variation at the earth's surface. MT method is widely used in exploration surveys worldwide, but it was hardly applicable in urban areas because of large artificial electromagnetic noise. But in one day, there are several hours at midnight that most of the electric equipment is shutdown. The MT time-series data at midnight is much quiet than in the daytime. Therefore, we focused on calculating the MT impedance using the quiet time-series data. In this research, the data observed by the Phoenix System is used. We introduced a robust impedance estimator based on the Hilbert-Huang transform (RMHHT) and used a new strategy to calculate the broadband MT time-series data. We indicated that this technique needs 4-hour time-series data to get a reliable resistivity structure up to 1,000 seconds in the numerical simulation. This short measurement time makes it possible to carry out MT surveys in urban areas with strong noises. However, the biggest problem of the impedance estimator based on HHT is time-consuming. The total computation time can be reduced significantly by using this strategy. Finally, a successful case study using the midnight time-series data was demonstrated to get reliable resistivity structures in the areas contaminated by heavy noises.

scholarly journals Transfer Matrix Computation for Wave Action Simulation in an Internal Combustion Engine

2012 ◽  
Author(s):  
Haitham Mezher ◽  
David Chalet ◽  
Pascal Chessé ◽  
Jérôme Migaud ◽  
Vincent Raimbault
Keyword(s):  
Frequency Domain ◽  
Mass Flow Rate ◽  
Transfer Matrix ◽  
Mass Flow ◽  
Flow Rate ◽  
Wave Action ◽  
Pressure Response ◽  
Series Data ◽  
Transfer Matrices ◽  
Dynamic Flow

A new technique for simulating engine pressure waves consisting of linking pressure response and mass flow rate excitation in the frequency domain has been presented. This is achieved on the so-called “dynamic flow bench”. With this new approach, precise, fast and robust results can be obtained while taking into account all the phenomena inherent to compressible unsteady flows. The method exhibited promising results when it was incorporated in a GT-Power/Simulink coupled simulation of a naturally aspirated engine. However, today’s downsized turbocharged engines come with more stringent simulation necessities, where discontinuities such as the charge air cooler (CAC) must be correctly modeled. Simulating such engines with the transfer function methodology is quite difficult because it requires mounting the entire intake line on the bench. Modeling wave action for these engines requires an understanding in the frequency domain of the flow’s characteristics through the different elements that make up the intake line. This leads us to study the acoustic transfer matrices. In order to split the intake line into separate elements, a straight duct of 185mm length is chosen as a first reference. It is mounted on the dynamic flow bench and pressure response is measured after an impulse mass flow excitation. Transfer functions of relative pressure and mass flow rate are then identified at given points upstream and downstream of this reference tube. These functions produce the desired transfer matrix poles. The resulting matrix is validated by inserting the tube in the intake lines of two four-cylinder engines which are modeled in GT-Power. Pressure and mass flow are registered at the measurement points of the tube from the simulation. The time series data upstream of the tube is treated in the frequency domain and the transfer matrix is used to calculate the corresponding downstream values. Measured values from the native simulation and those calculated using the transfer matrix propagation are then compared. Finally, the experimental technique for identifying transfer matrices of more complex elements using two versions of the previous tube is presented.

Short-Term Traffic Flow Forecasting for Urban Roads Using Data-Driven Feature Selection Strategy and Bias-Corrected Random Forests

2017 ◽  
Vol 2645 (1) ◽  
pp. 157-167 ◽  
Author(s):  
Jishun Ou ◽  
Jingxin Xia ◽  
Yao-Jan Wu ◽  
Wenming Rao
Keyword(s):  
Feature Selection ◽  
Traffic Flow ◽  
Random Forests ◽  
Urban Traffic ◽  
Data Driven ◽  
Series Data ◽  
Selection Strategy ◽  
Percentage Error ◽  
Short Term ◽  
Traffic Flow Forecasting

Urban traffic flow forecasting is essential to proactive traffic control and management. Most existing forecasting methods depend on proper and reliable input features, for example, weather conditions and spatiotemporal lagged variables of traffic flow. However, the feature selection process is often done manually without comprehensive evaluation and leads to inaccurate results. For that challenge, this paper presents an approach combining the bias-corrected random forests algorithm with a data-driven feature selection strategy for short-term urban traffic flow forecasting. First, several input features were extracted from traffic flow time series data. Then the importance of these features was quantified with the permutation importance measure. Next, a data-driven feature selection strategy was introduced to identify the most important features. Finally, the forecasting model was built on the bias-corrected random forests algorithm and the selected features. The proposed approach was validated with data collected from three types of urban roads (expressway, major arterial, and minor arterial) in Kunshan City, China. The proposed approach was also compared with 10 existing approaches to verify its effectiveness. The results of the validation and comparison show that even without further model tuning, the proposed approach achieves the lowest average mean absolute error and root mean square error on six stations while it achieves the second-best average performance in mean absolute percentage error. Meanwhile, the training efficiency is improved compared with the original random forests method owing to the use of the feature selection strategy.

scholarly journals Frequency–time analysis, low-rank reconstruction and denoising of turbulent flows using SPOD

2021 ◽  
Vol 926 ◽  
Author(s):  
Akhil Nekkanti ◽  
Oliver T. Schmidt
Keyword(s):  
Frequency Domain ◽  
Time Domain ◽  
Large Scale ◽  
Time Series Data ◽  
High Energy ◽  
Turbulent Jet ◽  
Low Rank ◽  
Series Data ◽  
Time Analysis ◽  
Frequency Domain Approach

Four different applications of spectral proper orthogonal decomposition (SPOD) are demonstrated on large-eddy simulation data of a turbulent jet. These are: low-rank reconstruction, denoising, frequency–time analysis and prewhitening. We demonstrate SPOD-based flow-field reconstruction using direct inversion of the SPOD algorithm (frequency-domain approach) and propose an alternative approach based on projection of the time series data onto the modes (time-domain approach). We further present a SPOD-based denoising strategy that is based on hard thresholding of the SPOD eigenvalues. The proposed strategy achieves significant noise reduction while facilitating drastic data compression. In contrast to standard methods of frequency–time analysis such as wavelet transform, a proposed SPOD-based approach yields a spectrogram that characterises the temporal evolution of spatially coherent flow structures. A convolution-based strategy is proposed to compute the time-continuous expansion coefficients. When applied to the turbulent jet data, SPOD-based frequency–time analysis reveals that the intermittent occurrence of large-scale coherent structures is directly associated with high-energy events. This work suggests that the time-domain approach is preferable for low-rank reconstruction of individual snapshots, and the frequency-domain approach for denoising and frequency–time analysis.

The reconstruction of the magnetotelluric impedance tensor: An adaptive parametric time‐domain approach

Geophysics ◽  
1988 ◽  
Vol 53 (8) ◽  
pp. 1080-1087 ◽  
Author(s):  
E. Yee ◽  
P. R. Kosteniuk ◽  
K. V. Paulson
Keyword(s):  
Frequency Domain ◽  
Time Domain ◽  
Auxiliary Information ◽  
Impedance Tensor ◽  
Adaptation Algorithm ◽  
Rational Form ◽  
On Line ◽  
Matrix Fraction ◽  
High Level ◽  
Domain Transformation

Magnetotelluric (MT) data processing for the estimation of the impedance tensor is usually carried out in the frequency domain using nonparametric methods of spectral analysis. In this paper, the impedance tensor is reconstructed using an adaptive parametric time‐domain approach, whereby the observed MT data are used directly in the estimation process without the need for a frequency‐domain transformation. In such a reconstruction, the impedance tensor is represented by a rational‐form or matrix‐fraction model. The parameters for this model are determined using a recursive instrumental variables (RIV) adaptation algorithm, allowing on‐line real‐time application. This adaptation algorithm is capable of providing consistent estimates for the impedance tensor from only the observed (i.e., noisy) MT field data and auxiliary information in the form of measurements of the contemporaneous components of the magnetic field at some remote site. Hence, the RIV algorithm provides a time‐domain implementation of the remote‐reference MT method, which has been applied with good success for unbiased impedance tensor determination in the frequency domain for moderate‐to‐high‐level noise conditions.

Sign in / Sign up

Export Citation Format

Share Document