Search

In order to improve the separation control effect of an oscillating jet, the external flow field of the actuators and the wing wake are obtained via hot-wire measurements to optimize the actuator and achieve wing separation flow control. The optimization objectives are to improve the sweeping uniformity and range of the jet. In the present study, the PCA method is used for the modal decomposition of the velocity distribution. The modal-based actuator evaluation parameters are proposed, and the kriging surrogate models of the modal coefficients (principal components) on the actuator parameters are established. The multi-objective genetic algorithm was utilized to complete the optimization of the actuator, and the effect of flow separation control on the wing was verified. The results show that three patterns exist in the time-averaged velocity distribution of the external flow field: unimodal, broad and bimodal, from unimodal to bimodal, the degree of the jet sweeping uniformity gradually decreases, and the sweeping range gradually increases. The pattern of the velocity distribution modals affects the degree of jet sweeping uniformity, while the distance of the modal peaks affects the jet sweeping range. The two evaluation parameters are negatively correlated: insufficient sweeping range or poor sweeping uniformity of the jet are not conducive to wing separation flow control, and the two must be coordinated to achieve the optimal control effect. [ABSTRACT FROM AUTHOR]

Spectral proper orthogonal decomposition(SPOD) has been recently used to capture temporal and spatial characteristics of complex flows. Applying SPOD analysis to such flow fields can deepen the understanding of flow mechanism. To further study the external flow at high Reynolds numbers, large eddy simulation(LES) is applied to acquire the flow field data while SPOD is utilized to extract important flow structures. The considered external flow includes incompressible flow around the cylinder and transonic shock oscillation. For the flow around the cylinder, flow field under three different Reynolds numbers is studied. Power spectral density(PSD) results of lift coefficient show that the vortex shedding frequency of low Reynolds number is S t = 0. 24 , which of moderate Reynolds number is S t = 0. 203 , and S t = 0. 27 for high Reynolds number flow. The increment of Reynolds number will lead to early instability of shear layer on the cylindrical surface. The modal results of SPOD and dynamic mode decomposition(DMD) at different Reynolds numbers show two antisymmetric structures at shedding vortex frequencies, indicating the resemblance in the process of vortex formation. For the shock wave/boundary layer interaction(SBLI) of transonic Royal Aircraft Establishment(RAE) 2822 airfoil, flow field under M ∞ = 0. 729 , R e c = 1. 94 × 1 0 7 is studied. The results show that the occurrence of unsteadiness is closely related to the normal shock wave/turbulent boundary layer interaction happening on the upper surface of the airfoil. Modal results of SPOD near the characteristic frequency show strong discontinuity near the shock wave. • Spectral proper orthogonal decomposition is developed to capture the spatiotemporal modes. • The mechanism of Reynolds-number effect on the flow around circular cylinder is revealed. • The low frequency of transonic buffet on airfoil is identified. [ABSTRACT FROM AUTHOR]

Cavitation is a complex hydrodynamic phenomenon that occurs when the local pressure in a liquid drops below the saturation vapor pressure, resulting in the phase transition of the liquid into vapor. This process was characterized by intricate three-dimensional flow structures and highly unsteady, violent dynamics. For applications such as hydraulic machinery, ship propulsion, and hydraulic engineering, cavitation often induces detrimental effects, and has significant impact on performance and reliability in engineering systems. In order to investigate the structure of non-constant cavitation flow field, this paper adopted the experimental method of image particle velocimetry PIV-LIF, which used a high-speed camera to photograph the cavitation flow from the side of the experimental section and followed the liquid flow by fluorescent tracer particles, so as to measure the cavitation velocity field. Using the high-resolution velocity field and grayscale snapshot data, the dynamic mode decomposition (DMD) was carried out to analyze the structure of the non-constant cavitation flow field. The DMD method can effectively decompose the flow field into several modes and identify the dominant mode for analysis. Based on the DMD method, the decomposition of the cavitated flow field has the characteristics of a single mode frequency and growth rate, which is more advantageous in the analysis of cavitated periodic flow. By analyzing the main modes that account for higher energy and performing modal reconstruction, it is found that all these modes contain certain characteristics of vacuole motion, which are related to the structure of the re-entrant jet and the accompanying vortex in the flow field, and the main evolution frequency of the vacuole is consistent with the vortex shedding frequency. The results of the flow field characteristics of each mode show that the first-order mode of the cavitation flow corresponds to a frequency of 0, which represents the average flow field; the second-order mode corresponds to a frequency of about the vacuole shedding frequency, which reveals that the vacuole grows and sheds periodically at the leading edge of the occurrence; the third-order mode corresponds to a frequency about twice that of the second-order mode, which reveals the fusion behavior of two large-scale vortices behind the hydrofoil; the fourth-order mode corresponds to a frequency about three times higher than the second-order mode, which characterizes the fusion behavior of some small-scale vortices in the flow field. Finally, the modal decomposition analysis of the cavitation flow field with different cavitation numbers was carried out, and it is found that the vortex structure of the shedding vacuole increases with decreasing cavitation number, and the second-order mode frequency decreases with decreasing cavitation number.

Microseismic monitoring data plays a crucial role in predicting and warning dynamic hazards such as coal mine rockburst. Utilizing historical microseismic monitoring data to forecast the evolution characteristics of future microseismic events is an effective approach to enhance the timeliness and accuracy of rockburst prediction and early warning. However, due to the non-stationary nature of microseismic data, conventional time series prediction methods often struggle to achieve accurate forecasts. To address this issue, this paper proposes a microseismic time series prediction method based on the combination of modal decomposition technique and deep learning methods. Firstly, the adaptive noise complete ensemble empirical mode decomposition (CEEMDAN) is employed to decompose the microseismic data into multiple intrinsic mode sequences. Subsequently, the decomposed sequences are reconstructed into high-frequency, low-frequency, and trend sequences using sample entropy. Furthermore, the variable mode decomposition (VMD) method is utilized to further decompose the high-frequency and low-frequency sequences into multiple new intrinsic mode sequences, which serve as the multivariate feature time series of microseismic data. Finally, a convolutional neural network (CNN) and long short-term memory network (LSTM) combined microseismic time series prediction model is established, with the microseismic multivariate feature time series within a certain time window and historical microseismic data as inputs, and the next moment microseismic monitoring data as output. The Bayesian optimization method is employed to optimize the hyperparameters of the model, aiming to improve the accuracy of microseismic prediction. Based on the microseismic monitoring data from the multiple working faces of the Xinjulong coal mine, the proposed method is applied to predict the daily maximum energy, daily average energy, and daily frequency of microseismic events, and a series of comparative experiments are conducted on the prediction of daily maximum energy data. The results show that the proposed method can effectively predict the evolution trend of microseismic events, with small errors between the predicted results and actual monitoring values. It demonstrates a good predictive performance and generalization ability.

Accurate wind speed prediction is extremely critical to the stable operation of power systems. To enhance the prediction accuracy, we propose a new approach that integrates bidirectional long short-term memory (BiLSTM) with fully adaptive noise ensemble empirical modal decomposition (CEEMDAN), the RIME optimization algorithm (RIME), and a multi-head self-attention mechanism (MHSA). First, the historical data of wind farms are decomposed via CEEMDAN to extract the change patterns and features on different time scales, and different subsequences are obtained. Then, the parameters of the BiLSTM model are optimized using the frost ice optimization algorithm, and each subsequence is input into the neural network model containing the MHSA for prediction. Finally, the predicted values of each component are weighted and reconstructed to obtain the predicted values of wind speed time series. According to the experimental results, the method can predict the short-term wind speeds of wind farms more accurately. We verified the effectiveness of the method by comparing it with different models. [ABSTRACT FROM AUTHOR]

In order to improve the safety of wind turbine, this paper takes the high-speed spur gear on output shaft in the transmission system of the wind turbine as the study object. The original signals of the gear in different fault states which obtained from the experimental platform decompose by the empirical mode decomposition method (EMD) and the variational mode decomposition method (VMD), respectively. Then the different gear fault types' eigenvectors were built. Fuzzy neural network (FNN) is adopted to learn the fault types and eigenvector samples of gears, and then the fault types of gears are identified. It is found that the fault features decomposed by VMD method have reached a high accurate recognition rate point to the fact that VMD has good applicability in the fault recognition of gear in wind turbine. [ABSTRACT FROM AUTHOR]

Pressure pulsations are crucial data within the flow field of a pump, and the characteristics of these pulsations can reveal changes in the internal flow. Based on model experiments, this paper obtained pressure pulsation data under two blade conditions and compared direct time-domain observations, peak-to-peak value changes, and the VMD decomposition method. The results show that even when it is known that one blade condition has changed, it is not possible to determine this through direct observation of pressure pulsation changes. The peak-to-peak value changes indicate that under special flow conditions, they are easily affected by different operating conditions, which can interfere with the results. In contrast, the VMD method, which decomposes the signal into low-frequency components, can better display anomalies within the pressure pulsation cycle and is less susceptible to the interference of flow conditions, offering some reference significance for diagnosing the blade operating conditions of the main pump. [ABSTRACT FROM AUTHOR]

Air pollution, particularly PM2.5, has long been a critical concern for the atmospheric environment. Accurately predicting daily PM2.5 concentrations is crucial for both environmental protection and public health. This study introduces a new hybrid model within the "Decomposition-Prediction-Integration" (DPI) framework, which combines variational modal decomposition (VMD), causal convolutional neural network (CNN), bidirectional long short-term memory (BiLSTM), and attention mechanism (AM), named as VCBA, for spatio-temporal fusion of multi-site data to forecast daily PM2.5 concentrations in a city. The approach involves integrating air quality data from the target site with data from neighboring sites, applying mathematical techniques for dimensionality reduction, decomposing PM2.5 concentration data using VMD, and utilizing Causal CNN and BiLSTM models with an attention mechanism to enhance performance. The final prediction results are obtained through linear aggregation. Experimental results demonstrate that the VCBA model performs exceptionally well in predicting daily PM2.5 concentrations at various stations in Taiyuan City, Shanxi Province, China. Evaluation metrics such as RMSE, MAE, and R2 are reported as 2.556, 1.998, and 0.973, respectively. Compared to traditional methods, this approach offers higher prediction accuracy and stronger spatio-temporal modeling capabilities, providing an effective solution for accurate PM2.5 daily concentration prediction. [ABSTRACT FROM AUTHOR]

Centrifugal pump pressure pulsation contains various signals in different frequency domains, which interact and superimpose on each other, resulting in characteristics such as intermittency, non-stationarity, and complexity. Computational Fluid Dynamics (CFD) and traditional time series models are unable to handle nonlinear and non-smooth problems, resulting in low accuracy in the prediction of pressure fluctuations. Therefore, this study proposes a new method for predicting pressure fluctuations. The pressure pulsation signals at the inlet of the centrifugal pump are processed using Variational Mode Decomposition–Particle Swarm Optimization (VMD-PSO), and the signal is predicted by Convolutional Neural Networks–Long Short-Term Memory (CNN-LSTM) model. The results indicate that the proposed prediction model combining VMD-PSO with four neural networks outperforms the single neural network prediction model in terms of prediction accuracy. Relatively high accuracy is achieved by the VMD-PSO-CNN-LSTM model for multiple forward prediction steps, particularly for a forward prediction step of 1 (Pre = 1), with a root mean square error of 0.03145 and an average absolute percentage error of 1.007%. This study provides a scientific basis for the intelligent operation of centrifugal pumps. [ABSTRACT FROM AUTHOR]

Short-term power load forecasting is the basis for ensuring the safe and stable operation of the power system. However, because power load forecasting is affected by weather, economy, geography, and other factors, it has strong instability and nonlinearity, making it difficult to improve the accuracy of short-term power load forecasting. To solve the above problems, a load forecasting method combining empirical modal decomposition (EMD) and long short-term memory neural network (LSTM) has been proposed. The original signal is first decomposed into a series of eigenmode functions and a residual quantity using the EMD algorithm. Subsequently, all the components are fed into the LSTM network. To further improve the load prediction accuracy, a self-attention mechanism is introduced for large component signals to further explore the internal correlation of the data, and the Sparrow Optimisation Algorithm (SSA) is used to optimize the LSTM hyperparameters. Combining EMD, LSTM, self-attention mechanism (SAM), and SSA, the EMD-SSASAM -LSTM method for short-term power load forecasting is further proposed. The results show that the coefficient of determination (R2) of the method is 0.98, the mean absolute error (MAE) is 0.013, the root mean square error (RMSE) is 0.018, and the mean absolute percentage error (MAPE) is 2.57%, which verifies that the proposed model can improve the accuracy of load forecasting, and has a certain application prospect. [ABSTRACT FROM AUTHOR]

In a “low-carbon” context, the power load is affected by the coupling of multiple factors, which gradually evolves from the traditional “pure load” to the generalized load with the dual characteristics of “load + power supply.” Traditional time-series forecasting methods are no longer suitable owing to the complexity and uncertainty associated with generalized loads. From the perspective of image processing, this study proposes a graphical short-term prediction method for generalized loads based on modal decomposition. First, the datasets are normalized and feature-filtered by comparing the results of Xtreme gradient boosting, gradient boosted decision tree, and random forest algorithms. Subsequently, the generalized load data are decomposed into three sets of modalities by modal decomposition, and red, green, and blue (RGB) images are generated using them as the pixel values of the R, G, and B channels. The generated images are diversified, and an optimized DenseNet neural network was used for training and prediction. Finally, the base load, wind power, and photovoltaic power generation data are selected, and the characteristic curves of the generalized load scenarios under different permeabilities of wind power and photovoltaic power generation are obtained using the density-based spatial clustering of applications with noise algorithm. Based on the proposed graphical forecasting method, the feasibility of the generalized load graphical forecasting method is verified by comparing it with the traditional time-series forecasting method.

The structural dynamic response reconstruction of large-scale civil engineering structures using traditional methods results in heavy computational burden with low efficiency. To improve efficiency, a novel time-domain response reconstruction method based on model condensation and modal decomposition is proposed in this paper. The finite element model of the target structure is transformed and condensed. The system matrices of the condensed model have lower dimensionality than those of the original finite element model. A checking procedure based on the modal assurance criterion is developed to ensure the accuracy of the condensed system. Measured dynamic responses are decomposed to time-domain modal responses at measured degrees of freedom (DOFs) using band-pass filters. The mode shape matrix of the condensed model is used to calculate the modal responses at the unmeasured DOFs. Following modal superposition and coordinate transformation, the dynamic responses at unmeasured DOFs are reconstructed without knowledge of the excitation. The accuracy and efficiency of the proposed methodology are verified in a numerical case study of a three-dimensional building frame model and an experimental case study of a two-story steel frame under laboratory conditions. In contrast with the numerical case study, the effects of modeling error and measurement noise are considered in the experiment verification. The results show the outstanding performance of the proposed methodology when compared with a traditional response reconstruction method. [ABSTRACT FROM AUTHOR]

In this work, a surrogate model for structural, transient and discontinuously excited finite element method simulations is developed. This allows to reduce the computational effort of repeated calculations of identical models under different load cases. The architecture of the surrogate combines fully connected neural network layers with long short-term memory layers. For the reproduction of different damping ratios, a categorical variable is added to the continuous input data. Based on a recursive flow of the predicted data back to the input layer, long-term dependencies do not vanish due to short-input sequences. The system dimension is reduced by applying the model-order reduction technique for modal decomposition. The high accuracy of the surrogate and the reduction of computational costs are shown on an academic example of a cantilever beam and a real-world example of a robot. The advantages of our approach are illustrated in comparison with state-of-the-art surrogates for transient finite element analysis. By using the surrogate proposed in this study, oscillations due to discontinuous excitation of mechanical structures can be reproduced. For this purpose, only short-input sequences are necessary since the excitation of the oscillations does not have to be part of the input sequence during the whole duration of the oscillations. Due to the categorical variable for the damping ratio, the surrogate can account for the influence of different damping in parameter studies. [ABSTRACT FROM AUTHOR]

The density of the ocean is unevenly distributed along the depth direction, showing a stratified structure. When there is an external disturbance, large-scale internal solitary waves are easily generated. The internal solitary waves are bounded by the intermediate pycnocline, and the currents in the upper and lower layers will flow in opposite directions. This generates strong shear forces that threaten the safety of marine structures. In this paper, the flow field distribution characteristics of a cylinder under the action of internal solitary waves at different scales are analyzed as a research object. The whole cylinder is discretized into 40 regions, and the horizontal force applied to each section of the cylinder is extracted. The force characteristics of the cylinder are analyzed. It is concluded that the pressure is the main factor determining the magnitude of the total combined force. In addition, the paper extracts the main flow structures from the modal decomposition point of view and explains the reasons affecting the force behavior of the cylinder. [ABSTRACT FROM AUTHOR]

Cavitation is a quasi-periodic process, and its non-stationarity leads to increasingly complex flow field structures. On the other hand, characterizing the flow field with greater precision has become increasingly feasible. However, accurately and effectively extracting the most representative vibration modes and spatial structures from these vast amounts of data has become a significant challenge. Researchers have proposed data-driven modal decomposition techniques to extract flow field information, which have been widely applied in various fields such as signal processing and fluid dynamics. This paper addresses the application of modal decomposition methods, such as dynamic mode decomposition (DMD), Proper Orthogonal Decomposition (POD), and Spectral Proper Orthogonal Decomposition (SPOD), in cavitation feature detection in hydraulic machinery. It reviews the mathematical principles of these three algorithms and a series of improvements made by researchers since their inception. It also provides examples of the applications of these three algorithms in different hydraulic machinery. Based on this, the future development trends and possible directions for the improvement of modal decomposition methods are discussed. [ABSTRACT FROM AUTHOR]

Pipelines constitute a vital component in offshore oil and gas operations, subjected to prolonged exposure to a range of alternating loads. Safeguarding their integrity, particularly through meticulous leak detection, is essential for ensuring safe and reliable operation. Acoustic emission detection emerges as an effective approach for monitoring pipeline leaks, demanding subsequent rigorous data analysis. Traditional analysis techniques like wavelet analysis, empirical mode decomposition (EMD), variational mode decomposition (VMD), and complete ensemble empirical mode decomposition with adaptive noise (CEEMDAN) often yield results with considerable randomness, adversely affecting leak detection accuracy. This study introduces an enhanced damage recognition methodology, integrating improved complete ensemble empirical mode decomposition with adaptive noise (ICEEMDAN) and probabilistic neural networks (PNN) for more accurate pipeline leak identification. This novel approach combines laboratory-acquired acoustic emission signals from leaks with ambient noise signals. Application of ICEEMDAN to these composite signals isolates eight intrinsic mode functions (IMFs), with subsequent time–frequency analysis providing insight into their frequency structures and feature vectors. These vectors are then employed to train a PNN, culminating in a robust neural network model tailored for leak detection. Conduct experimental research on pipeline leakage identification, focusing on the local structure of offshore platforms, experimental research validates the superiority of the ICEEMDAN–PNN model over existing methods like EMD, VMD, and CEEMDAN paired with PNN, particularly in terms of stability, anti-interference capabilities, and detection precision. Notably, even amidst integrated noise, the ICEEMDAN–PNN model maintains a remarkable 98% accuracy rate in identifying pipeline leaks. [ABSTRACT FROM AUTHOR]

In scenarios involving photovoltaic power generation and electrical dynamic loads, complex electricity metering signals often exhibit strong randomness and rapid fluctuations. These characteristics frequently lead to substantial errors in electricity metering, thereby affecting fair and equitable energy transactions. This paper presents a novel modal decomposition theory in the run-length domain. Such a theory is developed to map signals from the amplitude domain to run sequences in the run-length domain. Subsequently, it decomposes signals into a quasi-steady, slowly varying mode, and another dynamic, rapidly fluctuating mode, facilitating the extraction of sensitive characteristics over prolonged durations. Furthermore, the paper proposes characteristics representation method for complex electricity metering signals by constructing parameters and characteristic functions in the run-length domain. Additionally, the paper specifically extracts sensitive characteristics for photovoltaic complex electricity metering signals using the modal decomposition theory in the run-length domain. Finally, experimental validation is conducted to elucidate the impact of these sensitive characteristics on electricity meter errors with the maximum dynamic error observed at −15.53%.

In order to improve the separation control effect of an oscillating jet, the external flow field of the actuators and the wing wake are obtained via hot-wire measurements to optimize the actuator and achieve wing separation flow control. The optimization objectives are to improve the sweeping uniformity and range of the jet. In the present study, the PCA method is used for the modal decomposition of the velocity distribution. The modal-based actuator evaluation parameters are proposed, and the kriging surrogate models of the modal coefficients (principal components) on the actuator parameters are established. The multi-objective genetic algorithm was utilized to complete the optimization of the actuator, and the effect of flow separation control on the wing was verified. The results show that three patterns exist in the time-averaged velocity distribution of the external flow field: unimodal, broad and bimodal, from unimodal to bimodal, the degree of the jet sweeping uniformity gradually decreases, and the sweeping range gradually increases. The pattern of the velocity distribution modals affects the degree of jet sweeping uniformity, while the distance of the modal peaks affects the jet sweeping range. The two evaluation parameters are negatively correlated: insufficient sweeping range or poor sweeping uniformity of the jet are not conducive to wing separation flow control, and the two must be coordinated to achieve the optimal control effect.

Vortex shedding at the tail of a high-speed train changes the aerodynamic characteristics of the train, which affects the safety and stability of train operation. This paper takes CR400AF as the research object, and uses dynamic monitoring points to realize the whole process monitoring of the flow field at the tail of the train running in open air and in tunnel for the first time. The wake of the train in different infrastructure scenarios is analyzed by the proper orthogonal decomposition method. The study found that the wake vortex structure is quite different when the train runs in different scenarios, and the turbulent kinetic energy intensity of the wake in tunnel is higher than that of the open air running. Modal decomposition method can identify flow structures that have a large impact on train aerodynamics. Through frequency analysis, it is found that the modal frequency obtained from the decomposition is higher when running in open air than when running in tunnel. With the increase of train speed, the modal strouhal number increases when the train is running in open air, and decreases when the train runs in tunnel. After the train enters the tunnel, the reverse movement of the air around the train body suppresses the development and separation of the boundary layer, which is the main reason for the low frequency of wake vortex shedding in tunnel. The stability of the train running in tunnel is worse than that when running in open air, which is closely related to the more complex flow structure around the car body and the drastic change of aerodynamic force when running in the tunnel. [ABSTRACT FROM AUTHOR]

In this work, modal decomposition is employed to generate more data for matching scenarios with more complex topography for predicting wind power output in the case of complex terrain. The existing literature shows that a single wind power output forecast model is difficult to cope with complex terrain and thus the accuracy of wind power output forecast is limited. This work combines the advantages of attention mechanism and convolutional neural network for a novel network based on modal decomposition of historical data for wind power output forecast on complex terrain. The proposed novel network can break through the limitations of a single wind power output forecast model. In addition, the signals that are modally decomposed can be predicted more accurately. The presented method is contrasted with various other algorithms for the wind power output prediction problem in complex terrain. Comparative experiments show that the proposed network achieves a higher accuracy rate. [ABSTRACT FROM AUTHOR]

Short-term power load forecasting is the basis for ensuring the safe and stable operation of the power system. However, because power load forecasting is affected by weather, economy, geography, and other factors, it has strong instability and nonlinearity, making it difficult to improve the accuracy of short-term power load forecasting. To solve the above problems, a load forecasting method combining empirical modal decomposition (EMD) and long short-term memory neural network (LSTM) has been proposed. The original signal is first decomposed into a series of eigenmode functions and a residual quantity using the EMD algorithm. Subsequently, all the components are fed into the LSTM network. To further improve the load prediction accuracy, a self-attention mechanism is introduced for large component signals to further explore the internal correlation of the data, and the Sparrow Optimisation Algorithm (SSA) is used to optimize the LSTM hyperparameters. Combining EMD, LSTM, self-attention mechanism (SAM), and SSA, the EMD-SSA- SAM -LSTM method for short-term power load forecasting is further proposed. The results show that the coefficient of determination (R2) of the method is 0.98, the mean absolute error (MAE) is 0.013, the root mean square error (RMSE) is 0.018, and the mean absolute percentage error (MAPE) is 2.57%, which verifies that the proposed model can improve the accuracy of load forecasting, and has a certain application prospect.

This study presents a novel application of data-driven resolvent analysis algorithm for flow control. The objective is to identify key coherent structures connected to regions of the flow that are highly sensitive to structural changes. Modifying such regions, i.e., including momentum source terms, flow stabilizers, or changing the shape of the body under study, we can control the appearance of flow instabilities. The method is tested in two different applications: the laminar flow behind a two-dimensional circular cylinder and a turbulent channel flow including heat transfer to the endwall. In the first test case, the flow unsteadiness is controlled by including two stabilizers (formed by two small cylinders) in the areas suggested by resolvent analysis. This is a benchmark problem in fluid dynamics that serves to validate the idea of using this tool for flow control. In the second test case, data-driven resolvent analysis is applied as a mean to control, enhance or reduce, convective heat transfer. Modifications in the geometry, in the form of cavities and ribs included in the areas pointed by resolvent analysis, show the possibility of enhancing heat transfer while reducing drag. The findings highlight the importance of resolvent analysis in understanding flow dynamics and designing effective flow control strategies.

Accurate wind speed prediction is extremely critical to the stable operation of power systems. To enhance the prediction accuracy, we propose a new approach that integrates bidirectional long short-term memory (BiLSTM) with fully adaptive noise ensemble empirical modal decomposition (CEEMDAN), the RIME optimization algorithm (RIME), and a multi-head self-attention mechanism (MHSA). First, the historical data of wind farms are decomposed via CEEMDAN to extract the change patterns and features on different time scales, and different subsequences are obtained. Then, the parameters of the BiLSTM model are optimized using the frost ice optimization algorithm, and each subsequence is input into the neural network model containing the MHSA for prediction. Finally, the predicted values of each component are weighted and reconstructed to obtain the predicted values of wind speed time series. According to the experimental results, the method can predict the short-term wind speeds of wind farms more accurately. We verified the effectiveness of the method by comparing it with different models.

Pressure pulsations are crucial data within the flow field of a pump, and the characteristics of these pulsations can reveal changes in the internal flow. Based on model experiments, this paper obtained pressure pulsation data under two blade conditions and compared direct time-domain observations, peak-to-peak value changes, and the VMD decomposition method. The results show that even when it is known that one blade condition has changed, it is not possible to determine this through direct observation of pressure pulsation changes. The peak-to-peak value changes indicate that under special flow conditions, they are easily affected by different operating conditions, which can interfere with the results. In contrast, the VMD method, which decomposes the signal into low-frequency components, can better display anomalies within the pressure pulsation cycle and is less susceptible to the interference of flow conditions, offering some reference significance for diagnosing the blade operating conditions of the main pump.

Metal cutting is a complex process with strong randomness and nonlinear characteristics in its dynamic behavior, while tool wear or fractures will have an immediate impact on the product surface quality and machining precision. A combined prediction method comprising modal decomposition, multi-channel input, a multi-scale Convolutional neural network (CNN), and a bidirectional long-short term memory network (BiLSTM) is presented to monitor tool condition and to predict tool-wear value in real time. This method considers both digital signal features and prediction network model problems. First, we perform correlation analysis on the gathered sensor signals using Pearson and Spearman techniques to efficiently reduce the amount of input signals. Second, we use Complete ensemble empirical mode decomposition with adaptive noise (CEEMDAN) to enhance the local characteristics of the signal, then boost the neural network's identification accuracy. In addition, the deconstructed signal is converted into a multi-channel input matrix, from which multi-scale spatial characteristics and two-way temporal features are recovered using multi-scale CNN and BiLSTM, respectively. Finally, this strategy is adopted in simulation verification using real PHM data. The wear prediction experimental results show that, in the developed model, C1, C4, and C6 have good prediction performance, with RMSE of 8.2968, 12.8521, 7.6667, and MAE of 6.7914, 9.9263, and 5.9884, respectively, significantly lower than SVR, B-BiLSTM, and 2DCNN models. [ABSTRACT FROM AUTHOR]

Centrifugal pump pressure pulsation contains various signals in different frequency domains, which interact and superimpose on each other, resulting in characteristics such as intermittency, non-stationarity, and complexity. Computational Fluid Dynamics (CFD) and traditional time series models are unable to handle nonlinear and non-smooth problems, resulting in low accuracy in the prediction of pressure fluctuations. Therefore, this study proposes a new method for predicting pressure fluctuations. The pressure pulsation signals at the inlet of the centrifugal pump are processed using Variational Mode Decomposition–Particle Swarm Optimization (VMD-PSO), and the signal is predicted by Convolutional Neural Networks–Long Short-Term Memory (CNN-LSTM) model. The results indicate that the proposed prediction model combining VMD-PSO with four neural networks outperforms the single neural network prediction model in terms of prediction accuracy. Relatively high accuracy is achieved by the VMD-PSO-CNN-LSTM model for multiple forward prediction steps, particularly for a forward prediction step of 1 (Pre = 1), with a root mean square error of 0.03145 and an average absolute percentage error of 1.007%. This study provides a scientific basis for the intelligent operation of centrifugal pumps.

The density of the ocean is unevenly distributed along the depth direction, showing a stratified structure. When there is an external disturbance, large-scale internal solitary waves are easily generated. The internal solitary waves are bounded by the intermediate pycnocline, and the currents in the upper and lower layers will flow in opposite directions. This generates strong shear forces that threaten the safety of marine structures. In this paper, the flow field distribution characteristics of a cylinder under the action of internal solitary waves at different scales are analyzed as a research object. The whole cylinder is discretized into 40 regions, and the horizontal force applied to each section of the cylinder is extracted. The force characteristics of the cylinder are analyzed. It is concluded that the pressure is the main factor determining the magnitude of the total combined force. In addition, the paper extracts the main flow structures from the modal decomposition point of view and explains the reasons affecting the force behavior of the cylinder.

Cavitation is a quasi-periodic process, and its non-stationarity leads to increasingly complex flow field structures. On the other hand, characterizing the flow field with greater precision has become increasingly feasible. However, accurately and effectively extracting the most representative vibration modes and spatial structures from these vast amounts of data has become a significant challenge. Researchers have proposed data-driven modal decomposition techniques to extract flow field information, which have been widely applied in various fields such as signal processing and fluid dynamics. This paper addresses the application of modal decomposition methods, such as dynamic mode decomposition (DMD), Proper Orthogonal Decomposition (POD), and Spectral Proper Orthogonal Decomposition (SPOD), in cavitation feature detection in hydraulic machinery. It reviews the mathematical principles of these three algorithms and a series of improvements made by researchers since their inception. It also provides examples of the applications of these three algorithms in different hydraulic machinery. Based on this, the future development trends and possible directions for the improvement of modal decomposition methods are discussed.

Pipelines constitute a vital component in offshore oil and gas operations, subjected to prolonged exposure to a range of alternating loads. Safeguarding their integrity, particularly through meticulous leak detection, is essential for ensuring safe and reliable operation. Acoustic emission detection emerges as an effective approach for monitoring pipeline leaks, demanding subsequent rigorous data analysis. Traditional analysis techniques like wavelet analysis, empirical mode decomposition (EMD), variational mode decomposition (VMD), and complete ensemble empirical mode decomposition with adaptive noise (CEEMDAN) often yield results with considerable randomness, adversely affecting leak detection accuracy. This study introduces an enhanced damage recognition methodology, integrating improved complete ensemble empirical mode decomposition with adaptive noise (ICEEMDAN) and probabilistic neural networks (PNN) for more accurate pipeline leak identification. This novel approach combines laboratory-acquired acoustic emission signals from leaks with ambient noise signals. Application of ICEEMDAN to these composite signals isolates eight intrinsic mode functions (IMFs), with subsequent time–frequency analysis providing insight into their frequency structures and feature vectors. These vectors are then employed to train a PNN, culminating in a robust neural network model tailored for leak detection. Conduct experimental research on pipeline leakage identification, focusing on the local structure of offshore platforms, experimental research validates the superiority of the ICEEMDAN–PNN model over existing methods like EMD, VMD, and CEEMDAN paired with PNN, particularly in terms of stability, anti-interference capabilities, and detection precision. Notably, even amidst integrated noise, the ICEEMDAN–PNN model maintains a remarkable 98% accuracy rate in identifying pipeline leaks.

Electrical load forecasting is important to ensuring power systems are operated both economically and safely. However, accurately forecasting load is difficult because of variability and frequency aliasing. To eliminate frequency aliasing, some methods set parameters that depend on experiences. The present study proposes an adaptive hybrid model of modal decomposition and gated recurrent units (GRU) to reduce frequency aliasing and series randomness. This model uses average sample entropy and mutual correlation to jointly determine the modal number in the decomposition. Random adjustment parameters were introduced to the Adam algorithm to improve training speed. To assess the applicability and accuracy of the proposed hybrid model, it was compared with some state of the art forecasting methods. The results, which were validated by actual data sets from Shaanxi province, China, show that the proposed model had a higher accuracy and better reliability compared to the other forecasting methods. [ABSTRACT FROM AUTHOR]

Acoustic emission (AE) testing and Lamb wave inspection techniques have been widely used in non-destructive testing and structural health monitoring. For thin plates, the AEs arising from structural defect development (e.g., fatigue crack propagation) propagate as Lamb waves, and Lamb wave modes can be used to provide important information about the growth and localisation of defects. However, few sensors can be used to achieve the in situ wavenumber–frequency modal decomposition of AEs. This study explores the ability of a new multi-element piezoelectric sensor array to decompose AEs excited by pencil lead breaks (PLBs) on a thin isotropic plate. In this study, AEs were generated by out-of-plane (transverse) and in-plane (longitudinal) PLBs applied at the edge of the plate, and waveforms were recorded by both the new sensor array and a commercial AE sensor. Finite element analysis (FEA) simulations of PLBs were also conducted and the results were compared with the experimental results. To identify the wave modes present, the longitudinal and transverse PLB test results recorded by the new sensor array at five different plate locations were compared with FEA simulations using the same arrangement. Two-dimensional fast Fourier Transforms were then applied to the AE wavefields. It was found that the AE modal composition was dependent on the orientation of the PLB direction. The results suggest that this new sensor array can be used to identify the AE wave modes excited by PLBs in both in-plane and out-of-plane directions. [ABSTRACT FROM AUTHOR]

The direction of arrival (DOA) and number of sound sources is usually estimated by short-time Fourier transform and the conjugate cross-spectrum. However, the ability of a single AVS to distinguish between multiple sources will decrease as the number of sources increases. To solve this problem, this paper presents a multimodal fusion method based on a single acoustic vector sensor (AVS). First, the output of the AVS is decomposed into multiple modes by intrinsic time-scale decomposition (ITD). The number of sources in each mode decreases after decomposition. Then, the DOAs and source number in each mode are estimated by density peak clustering (DPC). Finally, the density-based spatial clustering of applications with the noise (DBSCAN) algorithm is employed to obtain the final source counting results from the DOAs of all modes. Experiments showed that the multimodal fusion method could significantly improve the ability of a single AVS to distinguish multiple sources when compared to methods without multimodal fusion. [ABSTRACT FROM AUTHOR]

The relation between Proper Orthogonal Decomposition (POD), Dynamic Mode Decomposition (DMD) and Flame Transfer Functions (FTF) is explored to gain further insight into the dynamics of two-dimensional laminar slit flames externally forced by velocity perturbations. The application of POD to the heat release rate fields suggests that the resultant modes can be split into two groups: the ones that are related to the displacement of the reactive sheet in the normal direction with regards to the flame front, and the ones that contain the local flame front distortions. The latter modes show preferential frequencies associated to the phase values of π , 2 π and 3 π of the FTFs, which can be related to the maximum gain value depending on the case. Furthermore, the results of the modal analysis seem to support that the flame tip dynamics can be conceptually modelled as a set of standing waves whose joint response can reconstruct a propagation in the normal flame front direction, generating also the temporal fluctuations of the spatially-integrated heat release in the domain. The DMD analysis shows the existence of an interaction between the flame and the flow, and illustrates the fundamental role played by the velocity perturbations at the base for the motion of the reactive sheet. Thus, the analysis shows how data-based decomposition methods can be used to identify complex physical phenomena contained in the FTF graphs with different levels of detail, and extend the modal analysis to the physical space. Novelty and significance statement This paper proves the potential of modal decomposition techniques like POD and DMD to infer the underlying physics of canonical flames and their relation to the FTF, validating also the hypotheses and methodology of other reduced-order models in the literature. A canonical configuration featuring a two-dimensional slit flame has been selected to analyse the flame dynamics using these methods. Hence, this work aims to set the foundations to elaborate data-driven reduced-order models based on modal decomposition techniques to support the analysis and the study of flame dynamics. [ABSTRACT FROM AUTHOR]

• A coordinated first-level voltage control framework with unified time scale is built. • Reactive power regulation problem by collecting the historical voltage is considered. • The empirical mode decomposition is embedded into real-time voltage controller. • Parameters of the designed controller are predicted by convolutional neural networks. • Fractional-order proportional-integral-derivative and transfer learning are combined. Extensive distributed energy connected to smart grids (SGs) poses a huge challenge to the voltage regulation of power systems. Conventional automatic voltage control (AVC) frameworks could lead to uncoordinated voltage control commands. To establish a standardized time scale, this study proposed a coordination primary voltage control (PVC) framework alongside a three-state energy (TSE) voltage regulation model. In this study, complete ensemble empirical mode decomposition with adaptive noise (CEEMDAN) is integrated with fractional-order proportional-integral-derivative (FOPID), alongside convolutional neural networks (CNNs) and transfer Q-learning (TQL) to form a unified approach termed C EEMDAN- FO PID- TQ L (CFOTQ). The CFOTQ aims to speed up the convergence of the controller and alleviate the lack of dynamic response in FOPID caused by fixed parameters. The proposed CFOTQ employs CEEMDAN to decompose voltage error sequences into implicit modal functions (IMFs) and a residual function (RF). FOPID parameters can be updated by CNNs. IMFs are controlled by the FOPID which provides accurate actions. RF controlled by TQL is skilled in dealing with high stochastic volatility and is characterized by migration capability. Remarkably, the errors of CFOTQ are reduced by 84.75 %, 83.82 %, and 88.72 % respectively compared to Q-learning, Q(λ), and R(λ). Consequently, the proposed CFOTQ is characterized by high control accuracy, small control error, and short training time. [ABSTRACT FROM AUTHOR]

We study the swing equation in the case of a multilayer network in which generators and motors are modeled differently; namely, the model for each generator is given by second order dynamics and the model for each motor is given by first order dynamics. We also remove the commonly used assumption of equal damping coefficients in the second order dynamics. Under these general conditions, we are able to obtain a decomposition of the linear swing equation into independent modes describing the propagation of small perturbations. In the process, we identify symmetries affecting the structure and dynamics of the multilayer network and derive an essential model based on a ‘quotient network.’ We then compare the dynamics of the full network and that of the quotient network and obtain a modal decomposition of the error dynamics. We also provide a method to quantify the steady-state error and the maximum overshoot error. Two case studies are presented to illustrate application of our method.

With advances in the scope of computational modeling methodologies, an increased focus is being placed on the application of data-driven techniques to increasingly complex problems. Due to the associated scale of the application, processing large datasets has emerged as a development bottleneck in practical applications of data-driven methods. While large-scale partial differential equation solvers are optimized for sparse linear algebra, many data-decomposition techniques (e.g. the singular value decomposition) require dense linear algebra operations. This work presents the tool PLATFORM which has enabled the application of modal decomposition and data-driven reduced-order modeling techniques for moderate (giga-) and large (tera-) scale data processing. The I/O and computing strategies and priorities are described. Most importantly, this framework uses abstraction techniques which allow users with limited understanding of distributed linear algebra computations and I/O to flexibly prototype and test methods on memory-intensive problems that are demanding in memory for the scripting environment.

In order to study the large scale flow structure of fire smoke in tunnels, this paper uses the excellent ability of proper orthogonal decomposition for the first time to extract flow field structure, processes the flow field data obtained from numerical simulation, obtains the 2-D and 3-D large-scale flow structure of fire smoke in tunnels and analyzes it. The proper orthogonal decomposition of temperature and velocity pulsation field shows that the proportion of first-order modal energy in temperature field is much higher than that in velocity field, indicating that the flow structure of velocity field is more complex than that of temperature field. Through modal phase analysis and modal decomposition of 3-D flow field, the influence of mode on fire smoke flow is recognized, and the understanding of tunnel fire smoke flow structure is deepened. The low-order mode is closely related to smoke flow. The increase of fire power has little effect on the reconstruction of flow field, and the cross-section energy recovery ratio near the fire source is higher under the same fire power. [ABSTRACT FROM AUTHOR]

The present work investigates turbulent flow structures and behavior near the leading edge of a longitudinal flow obstruction in an open channel with an inclined bank. A volumetric particle image velocimetry (VPIV) was employed to collect velocity data. The results indicate that a relatively moderate channel bank angle ( θ = 28 ∘ ) does not prevent the formation of a junction vortex (JV) system. Indeed, it is found that the JV system develops over the channel bank and extends to the leading edge of the flow obstruction. It is demonstrated that the main factor that strengthens the primary junction vortex (JV1) at the tip of the protrusion is the pronounced downward flow. The probability density functions (pdfs) of the velocity fluctuations specify that the JV1 oscillates aperiodically between the so-called zero-flow and back flow modes. This explains the amplification of TKE at its core. It is shown that the velocity difference between the peaks of the pdf increases from over the channel bank towards the tip of the flow obstruction. The aperiodic behavior of the JV1 is confirmed via the proper orthogonal decomposition (POD) technique. Specifically, it is demonstrated that the leading POD modes, associated with the JV system, contain lower energy content compared to periodic flows. The time-averaged vertical vorticity field verifies the presence of a tornado like vortex structure near the upstream face of the retaining wall. Finally, the results suggest that the largest bed shear stress values in the mean flow are located near the tip of the protrusion. [ABSTRACT FROM AUTHOR]

With the development of economy and the improvement of architectural aesthetics, civil structure buildings show a trend of diversification and complexity, which brings great challenges to the Structural Health Monitoring (SHM) of civil structure buildings. In order to optimise the structural health monitoring effect of civil structures, reduce monitoring costs, and improve the ability of civil structures to deal with risks, a civil structure health monitoring method combining Variational Modal Decomposition (VMD) and the Gated Recurrent Unit (GRU) is proposed. The gated neural network algorithm of modal decomposition is used, and then a wireless sensor network (WSN) civil structure health monitoring model is constructed on this basis. Finally, the application effect of the model is tested and analysed. The results show that the network energy consumption of this model can reach a minimum of 0.05 J, which is 0.05 J less than that of the Gate Recurrent Unit (GRU) model. The minimum loss value is 0.08. Its Mean Absolute Error (MAE), Root-Mean-Square Error (RMSE), and Mean Absolute Percent Error (MAPE) are 0.03, 0.04, and 0.06, respectively; the prediction error is the smallest, the overall amplitude difference monitored by the model remains at a low level of less than 0.01, and the changes are closest to the real situation. This shows that the model improves the operation efficiency, improves the accuracy of health monitoring, enhances the adaptability of building structural health monitoring to complex structures, provides a new way for the development of building structural health monitoring technology, and is conducive to enhancing civil structures. The safety and stability of buildings promote the high-quality development of civil and structural buildings. [ABSTRACT FROM AUTHOR]

The air-quality index (AQI) is an important comprehensive evaluation index to measure the quality of air, with its value reflecting the degree of air pollution. However, it is difficult to predict the AQI accurately by the commonly used WRF-CMAQ model due to the uncertainty of the simulated meteorological field and emission inventory. In this paper, a novel Auto-Modal network with Attention Mechanism (AMAM) has been proposed to predict the hourly AQI with a structure of dual input path. The first path is based on bidirectional encoder representation from the transformer to predict the AQI with the historical measured meteorological data and pollutants. The other path is a baseline to improve the generalization ability based on predicting the AQI by the WRF-CMAQ model. Several experiments were undertaken to evaluate the performance of the proposed model, with the results showing that the auto-modal network achieves a superior performance for all prediction lengths compared to some state-of-the-art models. [ABSTRACT FROM AUTHOR]

Accurate short-term multi-energy load forecasting is the cornerstone for optimal dispatch and stable operation of integrated energy system (IES). However, due to the complexity and coupling inside IES, multi-energy load forecasting faces serious challenges with data nonlinearity and instability, leading to reduced prediction accuracy. To this end, a novel short-term multi-energy load forecasting method for IES based on two-layer joint modal decomposition (TLJMD) and dynamic optimal ensemble (DOE) learning is developed in this paper. Firstly, the TLJMD method is proposed to decompose the nonlinear and nonstationary multi-energy load into several intrinsic mode functions (IMFs) to capture the periodicity and regularity within the multi-energy load. Secondly, the uniform information coefficient method is employed to select calendar, meteorological, and coupling feature that exhibit strong correlation with the multi-energy load. Eventually, the DOE model consisting of four base learners and the ensemble weight forecasting model is constructed, the IMFs and selected features are input into the DOE model to achieve the final forecasting results. The proposed method is tested on the publicly available data set from real-world scenario and compared with various forecasting methods to assess its effectiveness and accuracy. The simulation results indicate that the proposed method outperforms other forecasting methods in short-term multi-energy load forecasting for IES, with mean absolute percentage error values of 1.7025 %, 2.2244 %, and 2.3808 % for electric, heating, and cooling load forecasting, respectively. • A two-layer joint modal decomposition method is developed to capture the periodicity patterns within the load. • An external feature selection method based on uniform information coefficient is proposed innovatively. • A dynamic optimal ensemble strategy is developed, achieving the coordination and complementarity among base learners. • To verify the effectiveness of the proposed method, it is tested on the dataset from Arizona State University. [ABSTRACT FROM AUTHOR]

Based on the modal decomposition and least square method, full-depth tidal currents and corresponding internal tidal modes can be obtained from limited observations. Through a series of experiments based on direct observations at the mooring MP1 and synthetic observations at the moorings MP2 and MP3 in the South China Sea, the essential observations for reconstructing full-depth tidal currents are determined. Both observations in the upper ocean (especially near the thermocline) and deep layer (below 500 m) are essential in the reconstruction. For the upper ocean, an up-looking acoustic Doppler current profiler near the thermocline is generally essential, which can also be replaced by at least 2-4 current meters (CMs). The number of CM depends on the complexity and range of buoyancy frequency variation. For the deep layer, more observations are needed as water depth increases. The observations with the CM interval/water depth ratios about 0.2 can effectively reconstruct full-depth tidal currents below 500 m based on one barotropic mode and first three baroclinic modes. Significantly, the CMs in the middle of the deep layer are more important than those at two ends of the deep layer, which may provide more cost-effective design at a small loss of accuracy. Furthermore, the CMs with the depth interval/water depth ratios less than 1/8 in the deep layer can effectively reconstruct full-depth tidal currents with high-mode baroclinic motions.

The modal-domain structural model updating and damage detection methods own the advantage that they perform the whole computation without the information of system input, when compared to the time-domain. The system input required in the time-domain methods is hardly measured in realistic field tests of civil engineering structures. However, the time-domain methods own the advantage that they carry out model updating based on raw measured dynamic responses but not the processed (and compressed) modal parameters. The modal-domain methods only uses the compressed modal parameters, and some useful system information may be lost in the compressing process. Thus, higher errors and uncertainties are induced in the model updating results from the modal-domain methods. To take advantages and avoid disadvantages from the traditional time-domain and modal-domain model updating methods, a novel time-domain MCMC-based Bayesian model updating method without the information of the system input is developed in this paper by incorporating a newly developed dynamic response reconstruction technique and extending the formulation of the posterior probability density function (PDF) and the corresponding bridge PDF in the MCMC sampling process. A new modal decomposition algorithm is proposed in this paper for accurate modal response decomposition even in case of dense or sparse modes, and filters are not required for higher computation accuracy. Experimental case studies on a two-story steel frame were conducted to illustrate the proposed methodology in both structural model updating and damage detection. According to the analysis results, it is concluded that the proposed methodology outperforms the modal-domain methods especially in the cases with small number of measured DOFs. • No excitation information is needed with the proposed time-domain MCMC-based Bayesian model updating method. • A novel modal decomposition method is proposed to obtain the modal responses without filters. • The uncertainties introduced by some uncontrolled factors are quantified with the proposed technique. [ABSTRACT FROM AUTHOR]

Among the existing bearing fault diagnosis algorithms, the diagnosis time of algorithms with high accuracy is relatively long, and the accuracy of some lightweight methods is low compared to other methods. This work proposes a bi-level binary coded fully connected classifier based on residual network 50 with bottom and deep level features (Bi-BUR) for bearing fault diagnosis, enabling both high accuracy and high-speed characteristics. The Bi-BUR has two levels of classifiers. The first level begins with a signal-to-picture method based on integer binary coding to convert the bearing fault signals into binary pictures; then, the first classification of the converted pictures is performed by the underlying feature extraction-residual network 50, which can extract both the underlying and the advanced features. The second level performs modal decomposition of the signals that are not classified successfully in the first level, performs the second classification with a fully connected classifier inserted into the underlying feature extraction module, and exports the final classification results after weighted summation of the results of each type. The Bi-BUR is simulated on the Case Western Reserve University motor, the self-priming centrifugal pump, and the axial piston hydraulic pump bearing failure datasets. All three experiments achieved more than 95% accuracy in the first level of categorization and 100% diagnosis after the second level of categorization. And the Bi-BUR is faster than other high-accuracy methods. • A bearing fault diagnosis method with fusion of speed and accuracy is considered. • A two-layer bearing fault diagnosis method is proposed. • A signal-to-picture method based on integer binary coding is proposed. • A method to extract both bottom and deep level features is proposed. • The method fills the gap of bearing fault diagnosis with high speed and accuracy. [ABSTRACT FROM AUTHOR]

Ultra-long coal mine working faces are characterized by harsh conditions, making water supply pipelines susceptible to leakage incidents, posing risks to both personnel and equipment safety. However, detecting leaks using infrasound signals is challenging due to complex environmental noise interference. Additionally, spatial constraints make inspections difficult for maintenance personnel. Hence, it is vital to enhance the clarity of leakage data and accurately assess leak severity for effective pipeline maintenance. This study combines Complete Ensemble Empirical Mode Decomposition with Adaptive Noise (CEEMDAN) and Sparse Representation (SR) techniques to create a signal processing model. This model serves two purposes: reducing extraneous noise interference and enhancing leak-related signals. Subsequently, we analyse leak signal characteristics across various operational scenarios and formulate a leak severity assessment model using the Random Forest (RF) algorithm. Empirical experiments confirm its effectiveness, with a root mean square error (RMSE) of 0.1135 millimeters between predicted and actual leak severity values. This result validates the method's efficacy in estimating leak severity in water supply pipelines, offering a dependable approach to enhance pipeline maintenance and operations. • Leak detection utilizing infrasound waves with a longer propagation distance. • A joint denoising algorithm based on mode decomposition and sparse representation is proposed. • The introduction of a pipeline leakage prediction model, based on random forests, aims to describe the extent of pipeline leaks. [ABSTRACT FROM AUTHOR]

The deformation, flicker, and drift of a light field owing to complex media such as a turbulent atmosphere have limited its practical applications. Thus, research on invariants in randomly fluctuated light fields has garnered considerable attention in recent years. Coherence is a statistical property of light, while its full and quantitative characterization is challenging. Herein, we successfully realize the orthogonal modal decomposition of partially coherent beams and introduce the application of coherence entropy as a global coherence characteristic of such randomly fluctuated light fields. It is demonstrated that coherence entropy remains consistent during propagation in a unitary system by unraveling complex channels. As representative examples, we study the robustness of coherence entropy for partially coherent beams as they propagate through deformed optical systems and turbulent media. Coherence entropy is anticipated to serve as a key metric for evaluating the propagation of partially coherent beams in complex channels. This study paves the way for a broader application scope of a customized low-coherence light field through nonideal optical systems and complex media. [ABSTRACT FROM AUTHOR]