Wyniki wyszukiwania - BazTech

1

Short-term wind power combined prediction based on EWT-SMMKL methods

Li Jun, Ma Liancai

Archives of Electrical Engineering

|

2021

|

Vol. 70, nr 4

801--817

EN

Since wind power generation has strong randomness and is difficult to predict, a class of combined prediction methods based on empirical wavelet transform (EWT) and soft margin multiple kernel learning (SMMKL) is proposed in this paper. As a new approach to build adaptive wavelets, the main idea is to extract the different modes of signals by designing an appropriate wavelet filter bank. The SMMKL method effectively avoids the disadvantage of the hard margin MKL method of selecting only a few base kernels and discarding other useful basis kernels when solving for the objective function. Firstly, the EWT method is used to decompose the time series data. Secondly, different SMMKL forecasting models are constructed for the sub-sequences formed by each mode component signal. The training processes of the forecasting model are respectively implemented by two different methods, i.e., the hinge loss soft margin MKL and the square hinge loss soft margin MKL. Simultaneously, the ultimate forecasting results can be obtained by the superposition of the corresponding forecasting model. In order to verify the effectiveness of the proposed method, it was applied to an actual wind speed data set from National Renewable Energy Laboratory (NREL) for short-term wind power single-step or multi-step time series indirectly forecasting. Compared with a radial basic function (RBF) kernel- based support vector machine (SVM), using SimpleMKL under the same condition, the experimental results show that the proposed EWT-SMMKL methods based on two different algorithms have higher forecasting accuracy, and the combined models show effectiveness.

2

Seismic signal de noising using time–frequency peak fltering based on empirical wavelet transform

Liu Naihao, Yang Yang, Li Zhen, Gao Jinghuai, Jiang Xiudi, Pan Shulin

Acta Geophysica

|

2020

|

Vol. 68, no. 2

425--434

EN

Seismic noise suppression plays an important role in seismic data processing and interpretation. The time–frequency peak fltering (TFPF) is a classical method for seismic noise attenuation defned in the time–frequency domain. Nevertheless, we obtain serious attenuation for the seismic signal amplitude when choosing a wide window of TFPF. It is an unsolved issue for TFPF to select a suitable window width for attenuating seismic noise efectively and preserving valid signal amplitude efectively. To overcome the disadvantage of TFPF, we introduce the empirical wavelet transform (EWT) to improve the fltered results produced by TFPF. We name the proposed seismic de-noising workfow as the TFPF based on EWT (TFPFEWT). We frst introduce EWT to decompose a non-stationary seismic trace into a couple of intrinsic mode functions (IMFs) with diferent dominant frequencies. Then, we apply TFPF to the chosen IMFs for noise attenuation, which are selected by using a defned reference formula. At last, we add the fltered IMFs and the unprocessed ones to obtain the fltered seismic signal. Synthetic data and 3D feld data examples prove the validity and efectiveness of the TFPF-EWT for both attenuating random noise and preserving valid seismic amplitude.

3

An empirical wavelet transform based approach for multivariate data processing application to cardiovascular physiological signals

Singh O., Sunkaria R. K.

Bio-Algorithms and Med-Systems

|

2018

|

Vol. 14, no. 4

art. no. 20180030

EN

Background: This article proposes an extension of empirical wavelet transform (EWT) algorithm for multivariate signals specifically applied to cardiovascular physiological signals. Materials and methods: EWT is a newly proposed algorithm for extracting the modes in a signal and is based on the design of an adaptive wavelet filter bank. The proposed algorithm finds an optimum signal in the multivariate data set based on mode estimation strategy and then its corresponding spectra is segmented and utilized for extracting the modes across all the channels of the data set. Results: The proposed algorithm is able to find the common oscillatory modes within the multivariate data and can be applied for multichannel heterogeneous data analysis having unequal number of samples in different channels. The proposed algorithm was tested on different synthetic multivariate data and a real physiological trivariate data series of electrocardiogram, respiration, and blood pressure to justify its validation. Conclusions: In this article, the EWT is extended for multivariate signals and it was demonstrated that the component-wise processing of multivariate data leads to the alignment of common oscillating modes across the components.

4

Empirical wavelet transform-based delineator for arterial blood pressure waveforms

Singh O., Sunkaria R. K.

Bio-Algorithms and Med-Systems

|

2016

|

Vol. 12, no. 2

61--66

EN

Arterial blood pressure (ABP) waveforms provide plenty of pathophysiological information about the cardiovascular system. ABP pulse analysis is a routine process used to investigate the health status of the cardiovascular system. ABP pulses correspond to the contraction and relaxation phenomena of the human heart. The contracting or pumping phase of the cardiac chamber corresponds to systolic pressure, whereas the resting or filling phase of the cardiac chamber corresponds to diastolic pressure. An ABP waveform commonly comprises systolic peak, diastolic onset, dicrotic notch, and dicrotic peak. Automatic ABP delineation is extremely important for various biomedical applications. In this paper, a delineator for onset and systolic peak detection in ABP signals is presented. The algorithm uses a recently developed empirical wavelet transform (EWT) for the delineation of arterial blood pulses. EWT is a new mathematical tool used to decompose a given signal into different modes and is based on the design of an adaptive wavelet filter bank. The performance of the proposed delineator is evaluated and validated over ABP waveforms of standard databases, such as the MIT-BIH Polysomnoghaphic Database, Fantasia Database, and Multiparameter Intelligent Monitoring in Intensive Care Database. In terms of pulse onset detection, the proposed delineator achieved an average error rate of 0.11%, sensitivity of 99.95%, and positive predictivity of 99.92%. In a similar manner for systolic peak detection, the proposed delineator achieved an average error rate of 0.10%, sensitivity of 99.96%, and positive predictivity of 99.92%.