Wyniki wyszukiwania - Biblioteka Nauki

1

VMD and CNN-Based Classification Model for Infrasound Signal

100%

Lu Q. , Li M.

|

2023

|

tom Vol. 48, No. 3

403--412

EN

Infrasound signal classification is vital in geological hazard monitoring systems. The traditional classification approach extracts the features and classifies the infrasound events. However, due to the manual feature extraction, its classification performance is not satisfactory. To deal with this problem, this paper presents a classification model based on variational mode decomposition (VMD) and convolutional neural network (CNN). Firstly, the infrasound signal is processed by VMD to eliminate the noise. Then fast Fourier transform (FFT) is applied to convert the reconstructed signal into a frequency domain image. Finally, a CNN model is established to automatically extract the features and classify the infrasound signals. The experimental results show that the classification accuracy of the proposed classification model is higher than the other model by nearly 5%. Therefore, the proposed approach has excellent robustness under noisy environments and huge potential in geophysical monitoring.

2

Tool wear condition monitoring in milling process based on data fusion enhanced long short-term memory network under different cutting conditions

75%

Zheng G. , Sun W. , Zhang H. , Zhou Y. , Gao C.

|

tom Vol. 23, no. 4

612--618

EN

Tool wear condition monitoring (TCM) is essential for milling process to ensure the machining quality, and the long short-term memory network (LSTM) is a good choice for predicting tool wear value. However, the robustness of LSTM- based method is poor when cutting condition changes. A novel method based on data fusion enhanced LSTM is proposed to estimate tool wear value under different cutting conditions. Firstly, vibration time series signal collected from milling process are transformed to feature space through empirical mode decomposition, variational mode decomposition and fourier synchro squeezed transform. And then few feature series are selected by neighborhood component analysis to reduce dimension of the signal features. Finally, these selected feature series are input to train the bidirectional LSTM network and estimate tool wear value. Applications of the proposed method to milling TCM experiments demonstrate it outperforms significantly SVR- based and RNN- based methods under different cutting conditions.

3

Estimation of monthly evaporation values using gradient boosting machines and mode decomposition techniques in the Southeast Anatolia Project (GAP) area in Turkey

63%

Sarigöl M. , Katipoglu O. M.

|

tom Vol. 72, no. 2

999--1016

EN

Today, the biggest issue appears to be the increase in drought in some regions brought on by global warming, which has greatly increased the significance of water management. In light of evaporation's effect on drought, this research intends to evaluate the effectiveness of hybrid machine learning (ML) models, such as the Gradient Boosting Machines (GBM) technique paired with Empirical Mode Decomposition (EMD), Robust Empirical Mode Decomposition (REMD), Ensemble Empirical Mode Decomposition (EEMD), and Variational Mode Decomposition (VMD) signal decomposition techniques, for monthly evaporation prediction models in the Southeast Anatolia Project Area. In the design of the models, 80% of the data was used for training and 20% for testing. Furthermore, tenfold cross-validation was applied to solve the overfitting problem, which negatively affected the forecast performance. In the model setup, various combinations of precipitation, average air temperature, minimum air temperature, maximum air temperature, wind speed, actual air pressure, relative humidity, and solar time variables are presented to artificial intelligence models as input. The study revealed that the GBM methodology in combination with the signal decomposition methods REMD, EMD, EEMD, and VMD generally allowed for more accurate evaporation estimations than the GBM model alone. The study’s results are essential in relation to agricultural production, irrigation planning, water resources management studies, and hydrological modeling studies in the region.

4

An improved algorithm for efficient ocular artifact suppression from frontal EEG electrodes using VMD

63%

Dora C. , Biswal P. K.

|

tom Vol. 40, no. 1

148--161

EN

The Electroencephalogram (EEG) recordings from the frontal lobe of the human brain help in analyzing several important brain functions like motor functions, problem-solving skills, etc. or brain disorders. These recordings are often contaminated by high amplitude and long duration ocular artifacts (OAs) like eye blinks, flutters and lateral eye movements (LEMs), hence corrupting a considerable segment of EEG. In this study, an enhanced version of signal decomposition scheme i.e. Variational Mode Decomposition (VMD) based algorithm is used for suppression of OAs. The signal decomposition is preceded by identification of ocular artifact corrupted segment using Multiscale modified sample entropy (mMSE). The band limited intrinsic mode functions (BLIMFs) are obtained using predefined K (number of required BLIMFs) and α (balancing parameter). These parameters help to detrend the EEG segment in yielding the low frequency and high amplitude BLIMFs related to OA efficiently. Upon identifying OA components from the BLIMFs and estimating OA, it is regressed with the contaminated EEG to obtain the clean EEG. The proposed VMD based algorithm provides an improved performance in comparison with the existing single channel algorithms based on Empirical mode decomposition (EMD) and Ensembled EMD (EEMD) and multi-channel algorithms like Independent component analysis (ICA) and wavelet enhanced ICA for artifact suppression and is also able to overcome their limitations. The significance of the algorithm are: (1) no additional reference EOG channel requirement, (2) OA artifact based thresholds for identification and estimation from the mode functions obtained using VMD, and (3) also address the flutter artifacts.

5

Research on communication emitter identification based on semi-supervised dimensionality reduction in complex electromagnetic environment

51%

Ge W. , Qi L. , Tong L. , Zhu J. , Zhang J. , Zhao D. , Li K.

|

tom Vol. 71, nr 4

art. no. e145766

EN

The individual identification of communication emitters is a process of identifying different emitters based on the radio frequency fingerprint features extracted from the received signals. Due to the inherent non-linearity of the emitter power amplifier, the fingerprints provide distinguishing features for emitter identification. In this study, approximate entropy is introduced into variational mode decomposition, whose features performed in each mode which is decomposed from the reconstructed signal are extracted while the local minimum removal method is used to filter out the noise mode to improve SNR. We proposed a semi-supervised dimensionality reduction method named exponential semi-supervised discriminant analysis in order to reduce the high-dimensional feature vectors of the signals, and LightGBM is applied to build a classifier for communication emitter identification. The experimental results show that the method performs better than the state-of-the-art individual communication emitter identification technology for the steady signal data set of radio stations with the same plant, batch and model.