Improved Complete Ensemble Empirical Mode Decomposition With Adaptive Noise Research Articles

Missing interpolation model for wind power data based on the improved CEEMDAN method and generative adversarial interpolation network

Randomness and fluctuations in wind power output may cause changes in important parameters (e.g., grid frequency and voltage), which in turn affect the stable operation of a power system. However, owing to external factors (such as weather), there are often various anomalies in wind power data, such as missing numerical values and unreasonable data. This significantly affects the accuracy of wind power generation predictions and operational decisions. Therefore, developing and applying reliable wind power interpolation methods is important for promoting the sustainable development of the wind power industry. In this study, the causes of abnormal data in wind power generation were first analyzed from a practical perspective. Second, an improved complete ensemble empirical mode decomposition with adaptive noise (ICEEMDAN) method with a generative adversarial interpolation network (GAIN) network was proposed to preprocess wind power generation and interpolate missing wind power generation sub-components. Finally, a complete wind power generation time series was reconstructed. Compared to traditional methods, the proposed ICEEMDAN-GAIN combination interpolation model has a higher interpolation accuracy and can effectively reduce the error impact caused by wind power generation sequence fluctuations

Short-term wind speed forecasting based on a hybrid model of ICEEMDAN, MFE, LSTM and informer.

Wind energy, as a kind of environmentally friendly renewable energy, has attracted a lot of attention in recent decades. However, the security and stability of the power system is potentially affected by large-scale wind power grid due to the randomness and intermittence of wind speed. Therefore, accurate wind speed prediction is conductive to power system operation. A hybrid wind speed prediction model based on Improved Complete Ensemble Empirical Mode Decomposition with Adaptive Noise (ICEEMDAN), Multiscale Fuzzy Entropy (MFE), Long short-term memory (LSTM) and INFORMER is proposed in this paper. Firstly, the wind speed data are decomposed into multiple intrinsic mode functions (IMFs) by ICEEMDAN. Then, the MFE values of each mode are calculated, and the modes with similar MFE values are aggregated to obtain new subsequences. Finally, each subsequence is predicted by informer and LSTM, each sequence selects the one with better performance than the two predictors, and the prediction results of each subsequence are superimposed to obtain the final prediction results. The proposed hybrid model is also compared with other seven related models based on four evaluation metrics under different prediction periods to verify its validity and applicability. The experimental results indicate that the proposed hybrid model based on ICEEMDAN, MFE, LSTM and INFORMER exhibits higher accuracy and greater applicability.

An ICEEMDAN-based collaborative optimization control for wind-hydrogen-electrochemical energy storage under multiple application scenarios

An improved complete ensemble empirical mode decomposition with adaptive noise (ICEEMDAN)-based collaborative optimization control strategy of wind-hydrogen-electrochemical energy storage coupled system with the interconversion characteristics between hydrogen with electricity under multiple application scenarios is introduced in this paper. After identifying the grid-connected wind power based on the ICEEMDAN algorithm, the normalized maximum discrepancy of the modal functions divides the high-frequency modal components into the fluctuating power smoothed by lithium iron phosphate batteries and hydrogen storage, with wind power curtailment from grid connection being consumed by electrolysis of water for hydrogen in alkaline electrolyzers. Another novelty is a collaborative optimization strategy for hydrogen-electrochemical energy storage under two application scenarios, comparing the smoothing effect and the ability to eliminate wind curtailment with different energy storage schemes. Demonstrate the method's effectiveness through the certain operational data from a Chinese wind farm. Simulation results indicate that the coupled system results in 19.45% and 7.79% cost reduction compared to other schemes, and the collaborative optimization control strategy achieves complete wind curtailment, which further improves the capacity of consuming curtailed wind power while smoothing fluctuations and providing certain engineering application value.

A novel multivariable hybrid model to improve short and long-term significant wave height prediction

Accurate significant wave height (Hs) prediction is crucial for marine renewable energy development. The hybrid models combining multi-resolution analysis techniques such as empirical mode decomposition and wavelet transform with intelligence algorithm have flourished in Hs forecasting. However, these hybrid models cannot fit multivariable input mode well. In this study, a novel multivariable hybrid model is proposed. The improved complete ensemble empirical mode decomposition with adaptive noise (ICEEMDAN) and recurrence quantification analysis (RQA) were integrated as the deterministic and stochastic components decomposition (DSD) method. Then three machine learning models was integrated with DSD method as hybrid models, respectively. For more sufficient forecasting information, wind speed (Ws), wind direction (Wd) and Hs were adopted as inputs to construct multivariable hybrid models. The forecasting experiment was benchmarked with those from univariate hybrid models, multivariable single models and univariate single models. Three buoy-measured datasets were utilized for validation. Results revealed the positive effect of wind data on long-term prediction and the improvement to prediction by the DSD method. Benefiting from the advantages of both, multivariable hybrid models outperformed other benchmark models. Among them, the multivariable hybrid model based on long short-term memory (LSTM) network, DSD-LSTM-m, achieved the best forecasting performance.

Nonlinear Trend and Multiscale Variability of Dry Spells in Senegal (1951–2010)

Dry spells occurring during the rainy season have significant implications for agricultural productivity and socioeconomic development, particularly in rainfed agricultural countries such as Senegal. This study employs various chaos-theory-based tools, including the lacunarity method, rescaled analysis, and the improved complete ensemble empirical mode decomposition with adaptive noise (ICEEMDAN) method, to investigate the distribution, predictability, and multiscale properties of the annual series of maximum dry spell length (AMDSL) in Senegal during the rainy season. The analysis focuses on 29 stations across Senegal, spanning the period from 1951 to 2010. The findings reveal persistent behavior in the AMDSL across nearly all stations, indicating that predictive models based on extrapolating past time trends could enhance AMDSL forecasting. Furthermore, a well-defined spatial distribution of the lacunarity exponent β is observed, which exhibits a discernible relationship with rainfall patterns in Senegal. Notably, the lacunarity exponent displays a south-to-north gradient for all thresholds, suggesting its potential for distinguishing between different drought regimes and zones while aiding in the understanding of spatiotemporal rainfall variability patterns. Moreover, the analysis identifies five significant intrinsic mode functions (IMFs) characterized by different periods, including interannual, interdecadal, and multidecadal oscillations. These IMFs, along with a nonlinear trend, are identified as the driving forces behind AMDSL variations in Senegal. Among the inter-annual oscillations, a 3-year quasi-period emerges as the primary contributor and main component influencing AMDSL variability. Additionally, four distinct morphological types of nonlinear trends in AMDSL variations are identified, with increasing–decreasing and increasing trends being the most prevalent. These findings contribute to a better understanding of the variability in annual maximum dry spell lengths, particularly in the context of climate change, and provide valuable insights for improving AMDSL forecasting. Overall, this study enhances our comprehension of the complex dynamics underlying dry spell occurrences during the rainy season and presents potential avenues for predicting and managing the AMDSL in Senegal.

Comparative analysis of different empirical mode decomposition-kind algorithms on sea-level inversion by GNSS-MR

Abstract The rising sea level caused by global climate change might impact the human living environment. Global navigation satellite systems (GNSS)-multipath reflection (MR) technology holds significant potential for monitoring tide level changes. GNSS-MR technology typically employs low-order polynomials to extract the signal-to-noise ratio (SNR) residuals containing GNSS interference signals. It utilizes Lomb-Scargle (LSP) spectral analysis or empirical mode decomposition (EMD) to obtain the dominant frequency of the SNR residuals, which is then converted into tidal heights. However, as the satellite elevation angle increases, the GNSS interference signals decrease and the traditional method does not adapt well to the extraction of SNR residuals under such conditions. A series of improved EMD-kind algorithms, namely ensemble empirical mode decomposition (EEMD), complementary ensemble empirical mode decomposition (CEEMD), complete ensemble empirical mode decomposition with adaptive noise (CEEMDAN), and improved complete ensemble empirical mode decomposition with adaptive noise (ICEEDMAN), have been proposed to address the shortcomings of EMD algorithms such as end effect and mode aliasing. However, these improved EMD-kind algorithms have yet to be reported in sea level inversion. This study investigates the mitigation effects of EMD-kind algorithms on GNSS-MR direct signal and noise to improve the stability and accuracy of an SNR residual sequence with high satellite elevation angles. Experimental data from the HKQT station for one week and the SC02 station for one year are utilized to validate the effectiveness and accuracy of these algorithms in extracting SNR residuals. Compared to the traditional polynomial method, the experimental results demonstrate that all EMD-kind algorithms effectively address the distortion issue in traditional inversion methods under long periods, higher satellite elevation angles, and low GNSS receiver sampling rates. Among these algorithms, the results from the experiments show that ICEEMDAN consistently provides the best inversion accuracy. The results of the comparative analysis show that ICEEMDAN effectively reduces non-interference signals in SNR residuals at higher satellite elevation angles, expanding the useable range of satellite elevation angles and improving the utilization and temporal resolution of GNSS data inversion. Hence, it is an effective and appropriate approach to improving the accuracy of GNSS-MR tide level monitoring.

A novel fault location method based on traveling wave for multi-branch distribution network

In this paper, a novel fault location method is proposed for multi-branch distribution network based on traveling wave. The improved complete ensemble empirical mode decomposition with adaptive noise (ICEEMDAN) and Teager energy operator (TEO) is used to detect the arrival time of traveling waves. Firstly, according to the signal processing ability of the ICEEMDAN method, the ICEEMDAN method is used to decompose the sampled signals when the fault occurs, and extract the high-frequency component of the sampled signals. Then, the TEO is used to extract the abrupt characteristics of high-frequency modal components. And the first maximum mutation point of the signal is regarded as the initial fault traveling wave signal. In addition, considering the transmission characteristic of fault traveling wave signal in multi-branch lines of distribution network, the difference value characteristic of branches is defined. And the recognition principle and the fault location method of multi-branch lines are established to realize the accurate location of distribution network line faults. Simulation results show that the proposed ICEEMDAN-TEO method can effectively detect the fault traveling wave signals, and has a better detection performance than CEEMDAN-TEO and HHT methods. The proposed fault location method can effectively locate fault positions and reduce location errors.© 2017 Elsevier Inc. All rights reserved.

An active pole-to-ground fault location scheme for FB-MMC-MTDC system using the short-time blocking of sub-modules

The pole-to-ground (PTG) fault current is small in the multi-terminal direct current (MTDC) system, which cannot be used to locate the fault. This paper proposes a PTG fault location scheme for MTDC based on full bridge modular multilevel converter (FB-MMC), by actively blocking the sub-modules (SMs). The SMs blocking generates a voltage pulse wave which travels between the MMC and the fault point. By using the wave polarity analysis method, the reflected wave from fault point can be identified. The improved complete ensemble empirical mode decomposition with adaptive noise (ICEEMDAN) is used to detect the reflected wave arrival time. The time difference between pulse start time and reflection arrival time is processed for fault location. A four-terminal MMC-MTDC simulation model established in PSCAD/EMTDC validates the proposed method have not affected the power transmission. Afterwards, the fault location accuracy under the influences of fault type, fault location and fault impedance issues are also confirmed.

Research on SVM-Based Bearing Fault Diagnosis Modeling and Multiple Swarm Genetic Algorithm Parameter Identification Method

The bearing fault diagnosis of petrochemical rotating machinery faces the problems of large data volume, weak fault feature signal strength and susceptibility to noise interference. To solve these problems, current research presents a combined ICEEMDAN-wavelet threshold joint noise reduction, mutual dimensionless metrics and MPGA-SVM approach for rotating machinery bearing fault diagnosis. Firstly, we propose an improved joint noise-reduction method of an Improved Complete Ensemble Empirical Mode Decomposition with Adaptive Noise (ICEEMDAN) and wavelet thresholding. Moreover, the noise-reduced data are processed by mutual dimensionless processing to construct a mutual dimensionless index sensitive to bearing fault features and complete the fault feature extraction of the bearing signals. Furthermore, we design experiments on faulty bearings of multistage centrifugal fans in petrochemical rotating machinery and processed the input data set according to ICEEMDAN-wavelet threshold joint noise reduction and mutual dimensionless indexes for later validation of the model and algorithm. Finally, a support vector machine model used to effectively identify the bearing failures, and a multi-population genetic algorithm, is studied to optimize the relevant parameters of the support vector machine. The powerful global parallel search capability of the multigroup genetic algorithm is used to search for the penalty factor c and kernel parameter r that affect the classification performance of the support vector machine. The global optimal solutions of c and r are found in a short time to construct a multigroup genetic algorithm-support vector machine bearing fault diagnosis and identification model. The proposed model is verified to have 95.3% accuracy for the bearing fault diagnosis, and the training time is 11.1608 s, while the traditional GA-SVM has only 89.875% accuracy and the training time is 17.4612 s. Meanwhile, to exclude the influence of experimental data on the specificity of our method, the experimental validation of the Western Reserve University bearing failure open-source dataset was added, and the results showed that the accuracy could reach 97.1% with a training time of 14.2735 s, thus proving that the method proposed in our paper can achieve good results in practical applications.

A reliable fault detection algorithm for distribution network with DG resources

The natural disaster though may have less probability, but their impact can be extreme. For instance, windstorms can result in not only single instantaneous impact, but on multiple locations in the power grid. In such situations, detection of faults in active distribution networks (ADNs), having integration of different types of distributed generation (DG) resources, further escalates the challenging task for existing conventional techniques. The reliability of existing detection schemes is further compromised in case of high impedance faults (HIFs). This paper presents a unified approach for reliable detection of both low-impedance and high-impedance faults, while maintaining the discrimination against non-fault events. In the proposed detection algorithm, the fault detection index is calculated using a newer version of empirical mode decomposition (EMD), i.e., improved complete ensemble EMD with adaptive noise (ICEEMDAN). The performance is demonstrated for detection of multiple faults occurrence at different locations in the ADN. It is shown that with selection of uniform threshold value for all the relays in the network, only the dedicated primary and back-up relays operate reliably at the time of fault conditions. The scheme exhibits accurate, fast and reliable operation as compared to available works in the literature.

A novel model for runoff prediction based on the ICEEMDAN-NGO-LSTM coupling.

Prediction of runoff trends is a critical topic in hydrological forecasting. Accurate and reliable prediction models are important for the rational use of water resources. This paper proposes a new coupled model, ICEEMDAN-NGO-LSTM, for runoff prediction in the middle reaches of the Huai River. This model combines the excellent nonlinear processing capability of the Improved Complete Ensemble Empirical Mode Decomposition with Adaptive Noise (ICEEMDAN) algorithm, the perfect optimization strategy of the Northern Goshawk Optimization (NGO) algorithm, and the advantages of the Long Short-Term Memory (LSTM) algorithm in modeling time series data. The results show that the ICEEMDAN-NGO-LSTM model predicts the monthly runoff trend with higher accuracy compared to the actual data variation. The average relative error is 5.95% within 10%, and the Nash Sutcliffe (NS) is 0.9887. These results indicate that the ICEEMDAN-NGO-LSTM coupled model has superior prediction performance and provides a new method for short-term runoff forecasting.

Research on feature extraction method of spindle vibration detection of weak signals for rapier loom fault diagnosis in strong noise background

Fault diagnosis of rapier loom is an inevitable requirement to meet the demand of intelligent manufacturing. Facing the strong noise interference caused by complex working environment, accurate and reliable vibration signal detection of blade loom spindle is the key to realize the rapier loom fault diagnosis. This paper proposes a method to extract the spindle vibration signal of the rapier loom by Adaptive Piecewise Hybrid Stochastic Resonance (APHSR) after the Improved Complete Ensemble Empirical Mode Decomposition with Adaptive Noise (ICEEMDAN). Firstly, ICEEMDAN is used to pre-process the weak vibration signal containing noise, decompose the signal into multiple IMF components and display the high and low frequency signal characteristics of the original signal. Then, the energy density method and the correlation coefficient method are used to remove high and low noise, respectively, to filter the optimal IMF components, and then the signal containing valid information is reconstructed. Finally, the reconstructed signal is input to APHSR for noise-assisted enhancement after scale transformation to restore the faint vibration signal feature frequencies and achieve effective feature extraction. Through the simulation experiment and the engineering fault experiment analysis, comparing ICEEMDAN-APHSR with CEEMDAN-SR, ICEEMDAN-SR, CEEMDAN-APHSR methods. The difference between the spectrum amplitude, the spectrum amplitude and the maximum noise and the maximum signal to noise ratio (SNR) of the fault feature frequency of the rapier loom spindle bearing increased by 3.3668 dB,1.7205 dB,2.3952 dB, respectively. The results show that ICEEMDAN-APHSR method can accurately extract the fault feature frequency of the spindle bearing of rapier loom, and effectively solves the problem of extracting the weak vibration signal feature of rapier loom in the background of strong noise. This method is beneficial to the future research of rapier loom fault diagnosis, and is of great significance to promote the maintenance of loom equipment and production safety and quality.

Asymmetric information flow to G7 and Nordic equities markets during COVID-19 pandemic

PurposeThis paper investigates the probable differential impact of the confirmed cases of COVID-19 on the equities markets of G7 and Nordic countries to ascertain possible interdependencies, diversification and safe haven prospects in the era of the COVID-19 pandemic over the short-, intermediate- and long-term horizons.Design/methodology/approachThe authors apply a unique methodology in a denoised frequency-domain entropy paradigm to the selected equities markets (Li et al. 2020).FindingsThe authors’ findings reinforce the operability of the entrenched market dynamics in the COVID-19 pandemic era. The authors divulge that different approaches to fighting the pandemic do not necessarily drive a change in the deep-rooted fundamentals of the equities market, specifically for the studied markets. Except for an extreme case nearing the end (start) of the short-term (intermediate-term) between Iceland and either Denmark or the US equities, there exists no potential for diversification across the studied markets, which could be ascribed to the degree of integration between these markets.Practical implicationsThe authors’ findings suggest that politicians should pay closer attention to stock market fluctuations as well as the count of confirmed COVID-19 cases in their respective countries since these could cause changes to market dynamics in the short-term through investor sentiments.Originality/valueThe authors measure the flow of information from COVID-19 to G7 and Nordic equities using the entropy methodology induced by the Improved Complete Ensemble Empirical Mode Decomposition with Adaptive Noise (ICEEMDAN), which is a data-driven technique. The authors employ a larger sample period as a result of this, which is required to better comprehend the subtleties of investor behaviour within and among economies – G7 and Nordic geographical blocs – which largely employed different approaches to fighting the COVID-19 pandemic. The authors’ focus is on diverging time horizons, and the ICEEMDAN-based entropy would enable us to measure the amount of information conveyed to account for large tails in these nations' equity returns. Furthermore, the authors use a unique type of entropy known as Rényi entropy, which uses suitable weights to discern tailed distributions. The Shannon entropy does not account for the fact that financial assets have fat tails. In a pandemic like COVID-19, these fat tails are very strong, and they must be accounted for.

Multi-frequency information transmission among constituents and global equity returns: a sustainable and conventional way of investing

PurposeThis study investigates information flow of market constituents and global indices at multi-frequencies.Design/methodology/approachThe study’s findings were obtained using the Improved Complete Ensemble Empirical Mode Decomposition with Adaptive Noise (I-CEEMDAN)-based cluster analysis executed for Rényi effective transfer entropy (RETE).FindingsThe authors find that significant negative information flows among sustainability equities (SEs) and conventional equities (CEs) at most multi-frequencies, which exacerbates diversification benefits. The information flows are mostly bi-directional, highlighting the importance of stock markets' constituents and their global indices in portfolio construction.Research limitations/implicationsThe authors advocate that both SE and CE markets are mostly heterogeneous, revealing some levels of markets inefficiencies.Originality/valueThe empirical literature on CEs is replete with several dynamics, revealing their returns behaviour for diversification purposes, leaving very little to know about the returns behaviour of SE. Wherein, an avalanche of several initiatives on Corporate Social Responsibility (CSR) enjoin firms to operate socially responsible, but investors need to have a clear reason to remain sustainable into the foreseeable future period. Accordingly, the humble desire of investors is the formation of a well-diversified portfolio and would highly demand stocks to the extent that they form a reliable portfolio, especially, amid SEs and/or CEs.

Forecasting carbon price in China using a novel hybrid model based on secondary decomposition, multi-complexity and error correction

As global warming intensifies, the reduction of carbon emissions is imminent. Carbon price is directly related to whether carbon can be effectively reduced. Therefore, accurately forecasting carbon price has important practical significance. Aiming at the nonstationary and nonlinear characteristics of carbon price, this paper proposes a novel hybrid model for forecasting carbon price, which is based on improved complete ensemble empirical mode decomposition with adaptive noise (ICEEMDAN), multiscale fuzzy entropy (MFE), complete ensemble empirical mode decomposition (CEEMD), improved random forest by salp swarm algorithm (SSARF), improved back propagation by cuckoo search (CSBP), improved extreme learning machine by whale optimization algorithm (WOAELM) and error correction (EC), named ICEEMDAN-MFE-CEEMD-SSARF-CSBP-WOAELM-EC. Firstly, carbon price is decomposed by ICEEMDAN, divided into high-, medium-, and low-complexity components by MFE. Secondly, high-complexity components are merged and secondarily decomposed by CEEMD, which are still recorded as high-complexity components. Then, SSARF, CSBP and WOAELM are used to forecast high-, medium-, and low-complexity components, respectively, and forecasting results are reconstructed. Finally, EC is carried out using an extreme learning machine to obtain the final forecasting results, and the Diebold-Mariano test is introduced for a comprehensive evaluation of the model. Taking carbon price in the pilot cities of Shenzhen and Hubei as examples, after 6 aspects and 20 comparative experiments, the results show that the proposed model has higher forecast accuracy, with MAPE, MAE and RMSE up to 0.03131, 0.00089 and 4.02e-06 in Hubei, and its forecasting ability is better than other commonly used international carbon financial price forecasting models, providing a theoretical and data basis for carbon pricing and formulating carbon reduction policies in China. The main contributions of this paper are the improved primary decomposition, the use of secondary decomposition, and the innovative combination of three optimal models to forecast carbon price, but it still needs to be optimized for practice.

A new rainfall prediction model based on ICEEMDAN-WSD-BiLSTM and ESN.

Precipitation, as an important indicator describing the evolution of the regional climate system, plays an important role in understanding the spatial and temporal distribution characteristics of regional precipitation. Scientific and accurate prediction of regional precipitation is helpful to provide theoretical basis for relevant departments to guide flood and drought control. To address the uncertainty and nonlinear characteristics of precipitation series, this paper uses the established improved complete ensemble empirical mode decomposition with adaptive noise (ICEEMDAN)-wavelet signal denoising (WSD)-bi-directional long short-term memory (BiLSTM), and echo state network (ESN) models to predict precipitation of four cities in southern Anhui Province. The BiLSTM is used to predict the high-frequency components and the ESN to predict the low-frequency components, thus avoiding the influence between the two neural network predictions. The results show that the ICEEMDAN-WSD-BiLSTM and ESN models are more accurate. The average relative error reached 2.64% and the NSE (Nash-Sutcliffe efficiency coefficient) was 0.91, which was significantly better than the other four models. The model reveals the temporal change pattern and evolution characteristics of future precipitation, guides flood prevention and mitigation, and has certain theoretical significance and application value for promoting regional sustainable development.

Study of the Dynamical Relationships between PM2.5 and PM10 in the Caribbean Area Using a Multiscale Framework

The Caribbean basin is a geographical area with a high prevalence of asthma due to mineral dust. As such, it is crucial to analyze the dynamic behavior of particulate pollutants in this region. The aim of this study was to investigate the relationships between particulate matter with aerodynamic diameters less than or equal to 2.5 and 10 μm (PM2.5 and PM10) using Hilbert–Huang transform (HHT)-based approaches, including the time-dependent intrinsic correlation (TDIC) and time-dependent intrinsic cross-correlation (TDICC) frames. The study utilized datasets from Puerto Rico from between 2007 and 2010 to demonstrate the relationships between two primary particulate matter concentration datasets of air pollution across multiple time scales. The method first decomposes both time series using improved complete ensemble empirical mode decomposition with adaptive noise (ICEEMDAN) to obtain the periodic scales. The Hilbert spectral analysis identified two dominant peaks at a weekly scale for both PM types. High amplitude contributions were sustained for long and continuous time periods at seasonal to intra-seasonal scales, with similar trends in spectral amplitude observed for both types of PM except for monthly and intra-seasonal scales of six months. The TDIC method was used to analyze the resulting modes with similar periodic scales, revealing the strongest and most stable correlation pattern at quarterly and annual cycles. Subsequently, lagged correlations at each time scale were analyzed using the TDICC method. For high-frequency PM10 intrinsic mode functions (IMFs) less than a seasonal scale, the value of the IMF at a given time scale was found to be dependent on multiple antecedent values of PM2.5. However, from the quarterly scale onward, the correlation pattern of the PM2.5-PM10 relationship was stable, and IMFs of PM10 at these scales could be modeled by the lag 1 IMF of PM2.5. These results demonstrate that PM2.5 and PM10 concentrations are dynamically linked during the passage of African dust storms.

A Hybrid Deep Learning Model for Link Dynamic Vehicle Count Forecasting with Bayesian Optimization

The link dynamic vehicle count is a spatial variable that measures the traffic state of road sections, which reflects the actual traffic demand. This paper presents a hybrid deep learning method that combines the gated recurrent unit (GRU) neural network model with automatic hyperparameter tuning based on Bayesian optimization (BO) and the improved complete ensemble empirical mode decomposition with adaptive noise (ICEEMDAN) model. There are four steps in this hybrid approach. First, the ICEEMDAN is employed to decompose the link dynamic vehicle count time series data into several intrinsic components. Second, the components are predicted by the GRU model. At the same time, the Bayesian optimization method is utilized to automatically optimize the hyperparameters of the GRU model. Finally, the predicted subcomponents are reconstructed to obtain the final prediction results. The proposed hybrid deep learning method is tested on two roads of Hangzhou, China. Results show that, compared with the 12 benchmark models, the proposed hybrid deep learning model achieves the best performance in link dynamic vehicle count forecasting.

A new hybrid short-term carbon emissions prediction model for aviation industry in China

Given the China's construction of green aviation, a reliable carbon emissions prediction is essential to identify and mitigate the environmental impact of the aviation industry. In this regard, the current research aims at constructing “decomposition-prediction” to predict carbon emissions. To develop the above techniques, 1044 data points derived from the carbon emission accounts and datasets were used (70 % of the points for training and 30 % for validation). To solve the selection of mode number K and penalty factor α of variational mode decomposition (VMD) and calculate the matrix weight coefficient and kernel functions of extreme learning machine (ELM), optimized VMD and ELM is proposed. To begin with, improved complete ensemble empirical mode decomposition with adaptive noise (ICEEMDAN) is used to decompose the experimental data and spatial-dependence recurrence sample entropy (SdrSampEn) is used to quantify the complexity of all intrinsic mode functions (IMFs) to obtain the high and low complexity IMFs respectively. Then, high complexity IMFs are integrated, and then VMD of gray wolf optimizer algorithm (GWO) is used for decomposition to obtain IMFs, named G-IMFs. Next, low complexity IMFs are predicted by optimization ELM of slime mould algorithm (SMA) and G-IMFs are predicted by optimization ELM of particle swarm optimization algorithm (PSO). Lastly, all prediction results are reconstructed to obtain the final prediction results. To validate the superiority of the proposed method, 9 comparative models are established and their prediction effectiveness is assessed by R2 and RMSE. The results show that the proposed method has the best prediction performance for domestic aviation (R2 = 0.9978, RMSE = 0.0015) and international aviation (R2 = 0.9942, RMSE = 0.0003) respectively.

Application of ICEEMDAN to noise reduction of near-seafloor geomagnetic field survey data

Submarine geomagnetic field survey is a crucial means of geophysical exploration in the oceans. According to the potential field attenuation, it is preferable to carry out the near-seafloor magnetic survey close to the magnetic sources, for improving the resolution and accuracy of magnetic anomalies. However, due to the complicated marine environments and other factors, the real submarine magnetic data is often distorted by serious noises, which will result in the difficulty and low reliability of subsequent inversion and geological interpretation. Since the excellent adaptability and binary filtering characteristics, the Improved Complete Ensemble Empirical Mode Decomposition with Adaptive Noise (ICEEMDAN), revised from Empirical Mode Decomposition (EMD), has become a suitable choice for processing the nonlinear and non-stationary submarine magnetic data. Moreover, to effectively reconstruct the useful signal from the Intrinsic Mode Functions (IMFs) of ICEEMDAN, an improved adaptive shrinkage scheme related to interval extremum based on interval thresholding (IT) function is proposed. In this paper, the ICEEMDAN and adaptive IT techniques are combined and applied to denoise the near-seafloor geomagnetic field survey data. The synthetic tests show that, the proposed method has a better application effect for noise reduction than other methods. Meanwhile, the random noises, ocean wave interferences and gross errors can be effectively removed. The application to the really repeated marine magnetic survey profiles illustrates that the inner coincidence accuracy of the total magnetic intensity anomaly datasets along these two profiles, has been improved from ±16.750 nT to ±1.417 nT via data denoising. Both synthetic test and real data study suggest that proposed method has a good performance for denoising the near-seafloor geomagnetic field survey data.