Mode Decomposition Technique Research Articles

Short-Term Electricity Price Forecasting Using EEMD and GRU-NN

A vital necessity for the reliable operation of the power system as well as for the financial well-being of the consumers is an accurate forecast of the price of electricity. The price of electricity is highly volatile, nonlinear, and subject to seasonal fluctuations. The price series is also affected by electrical market uncertainty and demand, which is strongly reliant on weather and electricity usage time. This study presents a short-term electricity price forecasting method that follows the similar days approach comprising Ensemble Empirical Mode Decomposition technique. The suggested method utilizes a feature selection methodology that combines the Random Forest Regressor and the Gradient Boosting Regressor to determine which features are most important for the machine learning model. To forecast the electricity price, a Gated Recurrent Unit Neural Network (GRU-NN) is employed as the machine learning model. The GRU-NN has the capability of accurately capturing complicated temporal relationships in electricity price time series, which enables it to make correct predictions. To evaluate the validity of the proposed method, data from the PJM electricity market have been used. The simulation results demonstrate that the suggested method is superior to the existing technique, with significantly improved values achieved for both the mean absolute percentage error (MAPE) and the root mean square error (RMSE).

Adaptive modal identification of honeycomb thin-walled composite structures with pit defects under thermal modal testing using variational mode decomposition technique based on digital image correlation

The active imaging blue light high-speed three-dimensional digital image correlation (3D-DIC) system was combined with a transient aerodynamic heating system and random pulse excitation technology to establish a high-frequency thermal vibration optical test system capable of multi-temperature zone testing at 900 °C to accurately obtain the thermal modal parameters of thin-walled structures. The full-field temperature distribution of a novel honeycomb thin-walled composite structure (HTWCS) with pit defects under non-uniform temperature was accurately simulated through reinforcement learning. The first six modal frequencies of honeycomb thin-walled composite structure with pit defects at room temperature increased compared with the non-destructive honeycomb thin-walled composite structure, and the third and fourth modal shapes interchanged. An adaptive modal parameter identification method for full-field three-dimensional vibration measurements under high-temperature environments was developed by combining the successive multivariate variational mode decomposition (VMD) with the modal superposition method. This method can effectively identify close modes. The area ratio-based approach was used to evaluate the damping of the measured response. The finite element model (FEM) of the HTWCS with pit defects at different temperatures was updated using a multi-state step-by-step model updating method and the temperature-dependent material properties were established. The results showed that the introduction of a damping ratio improved the accuracy of the updated FEM of HTWCS with pit defects. The proposed technique is anticipated to provide an effective method for high-frequency thermal vibration optical measurement and modal identification of thin-walled structures under multi-temperature regions.

PQ event identification in PV-wind based distribution network with variational mode decomposition and novel feature enabled random forest classifier

Abstract The complexity raised due to the exponential growth of renewable power generators in modern power system evokes the necessity of detecting the potential power quality threats. With this context, the article proposes a novel power quality (PQ) detection method in PV and wind generator integrated distribution networks. In the first stage, the variational mode decomposition (VMD) technique is utilized to pull out the characteristic modes out of the voltage signal for nine different PQ events. In the later stage, a set of features are extracted from the decomposed signal to train a novel Feature Enabled Random Forest Classifier (FERFC). The feature enabling capability of Random Forest classifier is reckoned with the help of a single decision tree. It has been proved that the proposed novel VMD-FERFC has the ability to accurately identify and classify the PQ disturbances in the PV-wind-based distribution network. The performance of the suggested technique is verified in presence of noise. Further the supremacy of the method is established by comparing with other competitive reported PQ detection methods.

On erosion transition from the incubation stage to the accumulation stage in liquid impingement erosion

The evolution of surface profiles during erosion caused by the continuous impact of a liquid jet was studied using a water-jet apparatus. In this study, we focused on the erosion transition from the incubation stage to the accumulation stage during liquid impingement erosion, which is important for water-jet cleaning and aging management for nuclear power plants. By computing the local erosion rate distribution and surface roughness parameters, the formations of isolated pits and asperities were investigated. The structure of erosion caused by liquid impingement was extracted by using a dynamic mode decomposition technique. The surface profile changes were primarily caused by the formation of isolated pits and asperities during the incubation stage. The roughness parameters remained unchanged during the incubation stage, with barely measurable erosion depth, whereas the skewness of the surface profile began to increase during this stage. The initial erosion with a small-scale roughness gradually evolved into a large-scale crater. Dynamic mode decomposition analysis showed the presence of a short-wavelength (mode A) surface depression during the incubation stage, followed by a long-wavelength (mode B) erosion crater caused by the merging of mode A surface depressions during the accumulation stage.

Spatiotemporal mode decomposition of ultrashort pulses in linear and nonlinear graded-index multimode fibers

We develop a spatiotemporal mode decomposition technique to study the spatial and temporal mode power distribution of ultrashort pulses in long spans of graded-index multimode fiber, for different input laser conditions. We find that the beam mode power content in the dispersive pulse propagation regime can be described by the Bose–Einstein law, as a result of the process of power diffusion from linear and nonlinear mode coupling among nondegenerate mode groups. In the soliton regime, the output mode power distribution approaches the Rayleigh–Jeans law.

Variational mode decomposition enabled temporal convolutional network model for state of charge estimation

AbstractDue to the fast growth of electric vehicles (EVs) , estimation for Battery's State‐of‐charge (SOC) received significant research interests. The reason is that an accurate SOC estimation can significantly contribute to the reliability of EVs. A Variational Mode Decomposition (VMD) technique enabled Temporal Convolutional Network (TCN) model is proposed by the authors for SOC estimation. The proposed method first adopts time‐frequency analysis techniques to decompose voltage values into different frequency domains, each of which is analysed with the VMD technique to obtain its features as the input for the TCN model. Then, the proposed method combines outputs of different frequency domains with an attention module as the final output of the TCN model. Experiments on real battery datasets indicate that the proposed method outperforms the existing methods by 7.2% in mean absolute error and 6.13% in root mean square error. In addition, the error between the estimated and actual values using the proposed method is bounded by 2%.

Multi-modal IoT-based medical data processing for disease diagnosis using Heuristic-derived deep learning

Nowadays, the quick development of the Internet of Things (IoT) has changed our lifestyle. However, disease diagnosis is a difficult task owing to managing an enormous volume of data. Thus, in this research task, the latest IoT-assisted disease detection model for healthcare applications is developed. Initially, the patient data are taken from the IoT sensor devices in image, text, and signal data. The gathered patient data is further forward to the pre-processing phase. Accordingly, the images are gathered from patients’ data using IoT devices and forward to the feature extraction process. Feature extraction is done based on the IoT-collected images using VGG16 and Inception network. Moreover, the signals are garnered from the patient’s data and fed to the decomposition procedure. It is done by Empirical Mode Decomposition (EMD) technique. Then, the optimal feature selection of data, images, and signals is carried out by a hybrid algorithm termed Mating Probability-based Hybrid Strider Glowworm Swarm Optimization (MP-HSGSO). Finally, the Radial Basis Recurrent Neural Network (RBRNN) using MP-HSGSO is suggested to get the detected outcomes. Here, various brain diseases have been detected to classify the disease. This designed RBRNN method is utilized for efficient treatment, and also it provides the appropriate decision-making through processing various kinds of multimodal data. Throughout the simulation outcomes, the accuracy and precision of the offered method achieve 97 % and 96 %. Consequently, the simulation outcome of the offered method is proved that the designed model using a deep learning model shows enriched performance in clinical trial applications.

Dynamic modal characteristics of transverse mode instabilities in ytterbium-doped fiber laser oscillator

In recent years, transverse mode instability (TMI) has been widely observed in fiber laser amplifier systems. The transverse mode instability phenomenon in fiber laser oscillators is less studied. Here, we focus on the dynamical output properties, i.e., its temporal signal and modal characteristics in a 30-μm-core-diameter ytterbium (Yb)-doped fiber laser oscillator. The TMI occurs at a pumping power around 310 W. Different from amplifiers, the basic oscillation frequency is quite low, at around 100 Hz, changing with time and pump power. When the fiber laser oscillator operates beyond TMI threshold at 357 W or 377 W for a while, the temporal fluctuation slowly disappears together with a decreased oscillation frequency, and appears again later. Based on the mode decomposition technique, we find that during the period of fluctuation disappearance at 357 W, the power output stays low and the output beam is still a mixture of fundamental mode and higher-order modes. The fundamental mode content is calculated to be averagely higher when temporal fluctuation disappears, increasing from ∼57% to ∼63%. Our results indicate complex interaction between the fiber laser oscillation and the TMI effect, and calls for more attention into understanding TMI in fiber laser oscillators.

Data-driven modal decomposition methods as feature detection techniques for flow problems: A critical assessment

Modal decomposition techniques are showing a fast growth in popularity for their wide range of applications and their various properties, especially as data-driven tools. There are many modal decomposition techniques, yet Proper Orthogonal Decomposition (POD) and Dynamic Mode Decomposition (DMD) are the most widespread methods, especially in the field of fluid dynamics. Following their highly competent performance on various applications in several fields, numerous extensions of these techniques have been developed. In this work, we present an ambitious review comparing eight different modal decomposition techniques, including most established methods, i.e., POD, DMD, and Fast Fourier Transform; extensions of these classical methods: based either on time embedding systems, Spectral POD and Higher Order DMD, or based on scales separation, multi-scale POD (mPOD) and multi-resolution DMD (mrDMD); and also a method based on the properties of the resolvent operator, the data-driven Resolvent Analysis. The performance of all these techniques will be evaluated on four different test cases: the laminar wake around cylinder, a turbulent jet flow, the three-dimensional wake around a cylinder in transient regime, and a transient and turbulent wake around a cylinder. All these mentioned datasets are publicly available. First, we show a comparison between the performance of the eight modal decomposition techniques when the datasets are shortened. Next, all the results obtained will be explained in detail, showing both the conveniences and inconveniences of all the methods under investigation depending on the type of application and the final goal (reconstruction or identification of the flow physics). In this contribution, we aim at giving a—as fair as possible—comparison of all the techniques investigated. To the authors' knowledge, this is the first time a review paper gathering all these techniques have been produced, clarifying to the community what is the best technique to use for each application.

Numerical analysis of combustion noise in an atmospheric swirl-stabilized LDI burner through modal decomposition techniques

Combustion noise in gas turbine engines has recently become a relevant source of noise in the aircraft due to the appearance of new burner architectures that are intrinsically more unstable, and the optimization of other conventional noise sources in this mean of transport (e.g., jet, fan, airframe). In this work, a simulation setup for reactive conditions was prepared in the CONVERGE finite-volume package using the detailed chemistry SAGE solver to model the combustion of a benchmark case, which was solved using a LES approach with three different cell base sizes: 8,10,12 mm. A confined liquid-fueled swirl-stabilized burner located at the CORIA Laboratory, France, was used to validate the numerical results with the experimental measurements obtained at this facility. OH-PLIF measurements and PDA results for both phases were used to guarantee the accuracy of the numerical OH contours and the velocity profiles of both phases. These experimental measurements were collected at CORIA. After ensuring the stabilization of the numerical flame, the reactive simulations were extended with some adjustments in the time step to capture the acoustic motion. Several techniques like Fast Fourier Transform (FFT), Proper Orthogonal Decomposition (POD) and Dynamic Mode Decomposition (DMD) were used to analyze these results and confirm the presence of a Precessing Vortex Core (PVC) and a Vortex Breakdown Bubble (VBB) during the coupling of pressure, axial velocity and fuel mass fraction in reactive conditions. Furthermore, the acoustic analysis performed with a Helmholtz solver proved that the second longitudinal mode of the chamber (310 Hz) was present in the pressure signal (300 Hz in the LES calculations) and resonated with the Vortex Breakdown Bubble (VBB). However, this dominant frequency did not appear in the frequency distribution of the OH mass fraction and no feedback interaction between the acoustic and the combustion happened. Thus, only combustion noise was obtained.

Numerical Investigation of the Detonation Cell Bifurcation with Decomposition Technique

Bifurcation of the characteristic cells into multiple smaller cells and decay of those cells into single large characteristic cell is observed frequently. In the present study the bifurcation phenomenon of the detonation front is investigated for marginally unstable detonations using decomposition technique. Numerical analysis is carried out with detailed chemical kinetics for detonation propagation in H2/O2 mixtures at 10 kPa. The dynamic characteristics of the instability at the detonation front, such as the local oscillation frequency and the coherent spatial structure of the oscillation are also studied with dynamic mode decomposition (DMD) technique. The coherent structures of the primary and secondary detonation cells are analyzed during the cell bifurcation process and the mechanism in which the secondary cells are formed is investigated. It is demonstrated that the modal analysis categorizes the instability phenomena clearly and can be effectively utilized to identify the origin and source of the instability.

Numerical Study of Unstable Shock-Induced Combustion with Different Chemical Kinetics and Investigation of the Instability Using Modal Decomposition Technique

An unstable shock-induced combustion (SIC) case around a hemispherical projectile has been numerically studied which experimentally produced a regular oscillation. Comparison of detailed H2/O2 reaction mechanisms is made for the numerical simulation of SIC with higher-order numerical schemes intended for the use of the code for the hypersonic propulsion and supersonic combustion applications. The simulations show that specific reaction mechanisms are grid-sensitive and produce spurious reactions in the high-temperature region, which trigger artificial instability in the oscillating flow field. The simulations also show that specific reaction mechanisms develop such spurious oscillations only at very fine grid resolutions. The instability mechanism is investigated using the dynamic mode decomposition (DMD) technique and the spatial structure of the decomposed modes are further analyzed. It is found that the instability triggered by the high-temperature reactions strengthens the reflecting compression wave and pushes the shock wave further and disrupts the regularly oscillating mechanism. The spatial coherent structure from the DMD analysis shows the effect of this instability in different regions in the regularly oscillating flow field.

Low-frequency unsteadiness of recompression shock structures in the diffuser of supersonic ejectors

Supersonic ejectors are passive gasdynamic devices that compress a low-pressure fluid by utilizing the kinetic energy of a high-pressure fluid in a variable area duct. The ejector consists of a primary supersonic nozzle in a mixing duct where the secondary flow is entrained and mixed. The mixed flow can undergo a series of recompression shocks resulting in a subsonic flow in the diverging portion to aid pressure recovery. Recompression shocks usually lead to unsteady shock boundary layer interactions. The performance of the ejector is influenced by shear layers, shock and expansion waves, and their mutual interactions. While existing literature has extensively dealt with mixing of the primary and secondary flows, the unsteadiness in flow resulting from recompression shocks has been seldom investigated. Fluctuations in pressure due to the unsteadiness of the shock often lead structural fatigue issues. This paper reports a detailed investigation on low-frequency unsteadiness of recompression shock using high-speed schlieren images and dynamic pressure measurements. Modal analyses using proper orthogonal decomposition and dynamic mode decomposition techniques are used to determine the dominant spatial modes and associated frequencies. Multimodal frequencies ranging between 80 and 300 Hz are observed. These findings are further corroborated by Fourier and wavelet transformations of the experimentally measured wall static pressure signals. Subsequently, scaling parameter is established for the dominant frequencies based on flow velocities upstream of the shock and the distance between two consecutive shocks. This results in a unique scaling frequency of 4.58% ± 18%, for the recompression shock independent of operating conditions.

Machine Learning Enabled P300 Classifier for Autism Spectrum Disorder Using Adaptive Signal Decomposition

Joint attention skills deficiency in Autism spectrum disorder (ASD) hinders individuals from communicating effectively. The P300 Electroencephalogram (EEG) signal-based brain-computer interface (BCI) helps these individuals in neurorehabilitation training to overcome this deficiency. The detection of the P300 signal is more challenging in ASD as it is noisy, has less amplitude, and has a higher latency than in other individuals. This paper presents a novel application of the variational mode decomposition (VMD) technique in a BCI system involving ASD subjects for P300 signal identification. The EEG signal is decomposed into five modes using VMD. Thirty linear and non-linear time and frequency domain features are extracted for each mode. Synthetic minority oversampling technique data augmentation is performed to overcome the class imbalance problem in the chosen dataset. Then, a comparative analysis of three popular machine learning classifiers is performed for this application. VMD's fifth mode with a support vector machine (fine Gaussian kernel) classifier gave the best performance parameters, namely accuracy, F1-score, and the area under the curve, as 91.12%, 91.18%, and 96.6%, respectively. These results are better when compared to other state-of-the-art methods.

Comparison of empirical modal decomposition class techniques applied in noise cancellation for building heating consumption prediction based on time-frequency analysis

Empirical Modal Decomposition (EMD), and improved or modified techniques derived from EMD, collectively referred to as Empirical Modal Decomposition class (EMDC) techniques. EMDC techniques have a wide range of applications in building energy analysis, especially time–frequency analysis based noise cancellation in data-driven building energy prediction. However, there is a gap in the literature related to the choice of EMDC techniques in data-driven models. This paper provides a framework for a comprehensive comparison of EMD, Ensemble Empirical Mode Decomposition (EEMD), Variational Mode Decomposition (VMD) and Empirical Wavelet Transform (EWT) techniques for building heat consumption prediction modeling. A real building is used as an example to compare the noise cancellation potential of these techniques and the prediction accuracy under various data-driven models. The results demonstrated that noise cancellation using the EMDC techniques significantly improves Signal-Noise Ratio, regularity, and consistency with the original signal trend. The prediction models trained using the noise-cancelled data have the Root Mean Squared Error and the Mean Absolute Error reductions of 22.5 % and 31.3 % on average, respectively. Meanwhile, the predicted signals of the models inherit the noise cancellation benefits of the noise-cancelled training data.

Modal Decomposition Analysis of Bluff-Body Stabilized Lean Premixed CH4/H2/air Flames Based on LES Data

Unsteady flow and flame dynamics of bluff body stabilized CH4/H2/air premixed flames are investigated numerically with LES and modal decomposition techniques. Energy ranked and frequency dependent modes are identified with proper orthogonal (POD), spectral proper orthogonal (SPOD) and dynamic mode decompositions (DMD). Flow and flame coherent structures are extracted based on the axial velocity and heat release rate contours. Two cases, namely 100% CH4/air (V-flame) and 43.4% CH4 + 56.6% H2/air (M-flame), are selected for the analysis. Each case is acoustically excited in large eddy simulation (LES) with harmonic excitation signals at distinct frequencies and with a broadband excitation signal, to produce snapshots for modal decomposition analysis. In POD, the frequencies of the relevant modes are extracted with the discrete Fourier transform (DFT) of time coefficients, while in SPOD and DMD the extracted modes are frequency dependent by nature. Results are compared and commented.

Study of nonlinear characteristics in hollow cathode discharge in unmagnetized plasmas

AbstractParametric study and nonlinear dynamical analysis of the discharge parameter variations are presented here for a hollow cathode source in the glow discharge regime and later in the radio frequency (RF) discharge regime. A plasma source consisting of a semi‐transparent hollow cathode leads to the generation of nonthermal energetic charged particles and the formation of multiple potential regimes in the plasma column during both the glow discharge as well as the RF discharge regimes. The hollow cathode discharge obtained using this single plasma source exhibits significant mixed‐mode oscillations (MMOs) during the glow discharge regime. Various regular and irregular sequences of MMOs evolved during the progressive change of discharge voltage in the glow discharge regime. During the RF power application across the hollow cathode source, the charged particles oscillate and appear in bursts in the time series. The present study investigates energy distribution among the coherent modes of discharge through the application of the Empirical Mode Decomposition technique. Multiple coherent modes are identified during both discharge regimes that carry most of the discharge energy.

Automatic modal identification of bridges based on free vibration response and variational mode decomposition technique

Free vibration tests are often performed for the dynamic characterization of structural systems and components. In the field of road bridges monitoring, such tests require minimal equipment (i.e., a sparse network of few sensors, even a single one) in order to gather – quickly and at low cost – some essential information about the dynamic behavior of the structure. The experimental free-vibration response is thus processed for estimating structural modal features which, in turn, can be useful to corroborate or update available numerical models as well as to evaluate current bridge conditions. Within this framework, an efficient computational strategy is here developed for the modal identification of bridges based on their free vibration response. The procedure combines the variational mode decomposition technique with the area ratio-based damping identification method, which ensures enhanced robustness to noise-related effects in field-test signals. A novel procedure is implemented to make the whole identification process fully automatic, in such a way that it does not depend on specific structural features (e.g., to identify even closely-spaced modes) or subjective user’s settings. After a preliminary validation with a synthetic multi-modal signal, the proposed procedure is applied to two real case-studies. The first case-study deals with the identification of natural frequencies, modal damping ratios and mode shapes of prestressed concrete girder bridge decks. Particularly, experimental natural frequencies and mode shapes are compared with numerical predictions obtained from a finite element model of the bridge. The dynamic characterization of the cables of a cable-stayed bridge is considered in the second case-study. Herein, experimental natural frequencies of all the stay-cables are initially analyzed for estimating their stress levels, which have been critically analyzed to detect the existence of relaxation losses. Moreover, modal damping ratios of all the stay-cables are identified and compared to estimates obtained by alternative ambient-vibration-based techniques.

Transonic Shock Buffet Flowfield Assessment Using Various Dynamic Mode Decomposition Techniques

Dynamic mode decomposition (DMD) is a powerful data-driven modal decomposition technique that extracts spatiotemporal coherent structures: a useful process in flow diagnostics and future state estimation of complex nonlinear flow phenomena. Transonic shock buffet is a complicated phenomenon, and modal decomposition techniques such as DMD provide significant insight into its complicated flow physics; but, often, flowfield data are corrupted because of various sources of noise due to the presence of outliers or the absence of critical data components. Therefore, noise corruption renders the modal decomposition inaccurate, and thereby not useful. In this paper, two sources of noise have been considered: simple white noise, and complex salt-and-pepper-type spurious noise. Various DMD techniques including standard DMD, forward–backward DMD, total-least-squares DMD, higher-order DMD, and robust DMD have been implemented. Their effectiveness and limitations in countering noise corruption have been investigated systematically. In the case of white noise corruption, forward–backward DMD, total-least-squares DMD, and higher-order DMD capture the buffet frequency and growth rate with sufficient accuracy, whereas the latter outperforms the other two when the noise variance level is above 5%. In the case of spurious noise, robust DMD handles noise corruption efficiently, with surprisingly high values of pixel corruption of up to 30%.

Short-term PV power forecasting based on time series expansion and high-order fuzzy cognitive maps

Achieving short-term accurate photovoltaic power forecasting is of great significance to improve the efficiency of grid operation, especially in power stations where historical values of meteorological parameters are lagged or not recorded. Therefore, numerous studies have been presented to predict the intricate stochastic properties of photovoltaic power sequence in recent years. However, due to the limitation of algorithm performance and the effect of noise in the time series, previously proposed methods may not extract effective features from power data. Furthermore, most studies have only focused on point prediction, which ignores the uncertain information and unavoidable forecast bias. In this study, a hybrid framework based on fuzzy information granulation algorithm, improved variational mode decomposition technique, and high-order fuzzy cognitive maps is proposed to fill these gaps. Comparison experiments were set up using 5-minute photovoltaic power data from Alice Springs, Australia. The computational results not only demonstrate that the proposed framework significantly improves forecast accuracy of short-term photovoltaic power, but also achieves effective interval prediction by fuzzy information.