Time-dependent Features Research Articles

An RFE-aided Transformer-SVM framework for multi-bolt connection loosening identification using wavelet entropy of vibro-acoustic modulation signals

To ensure structural safety and integrity, a novel framework is developed for detecting the loosening of multi-bolt connections using wavelet entropy of vibro-acoustic modulation (VAM) signals. Wavelet entropy is employed as the dynamic index to capture the intricate time-frequency characteristics that are indicative of the connection status. Taking the wavelet entropy vectors as input, the proposed framework distinguishes itself by integrating a Transformer model for high-dimensional feature extraction with the recursive feature elimination (RFE) for essential feature selection, followed by a support vector machine (SVM) model for classification. Specifically, the Transformer model with innovative positional encoding capability helps to extract the time-dependent transient features that are sensitive to the bolt loosening. The RFE process reduces the data dimensionality while discerning the diagnostic information for more accurate classification. Through the experiment on a four-bolt joint, the identification results with cross-validation showed high accuracy and robustness of the proposed framework across various loosening cases. It outperformed the traditional SVM, long short-term memory network (LSTM), convolutional neural network (CNN)-SVM models without and with RFE, as well as the Transformer-SVM model without RFE, achieving an accuracy increase of 15.72%, 11.74%, 9.47%, 5.49%, and 5.06%, respectively. The proposed framework was demonstrated to be able to learn the damage-sensitive features more effectively from wavelet entropy data, marking a significant advancement in the health monitoring of engineering structures with high-strength bolt connections.

Deep learning for predicting rehospitalization in acute heart failure: Model foundation and external validation.

Assessing the risk for HF rehospitalization is important for managing and treating patients with HF. To address this need, various risk prediction models have been developed. However, none of them used deep learning methods with real-world data. This study aimed to develop a deep learning-based prediction model for HF rehospitalization within 30, 90, and 365days after acute HF (AHF) discharge. We analysed the data of patients admitted due to AHF between January 2014 and January 2019 in a tertiary hospital. In performing deep learning-based predictive algorithms for HF rehospitalization, we use hyperbolic tangent activation layers followed by recurrent layers with gated recurrent units. To assess the readmission prediction, we used the AUC, precision, recall, specificity, and F1 measure. We applied the Shapley value to identify which features contributed to HF readmission. Twenty-two prognostic features exhibiting statistically significant associations with HF rehospitalization were identified, consisting of 6 time-independent and 16 time-dependent features. The AUC value shows moderate discrimination for predicting readmission within 30, 90, and 365days of follow-up (FU) (AUC:0.63, 0.74, and 0.76, respectively). The features during the FU have a relatively higher contribution to HF rehospitalization than features from other time points. Our deep learning-based model using real-world data could provide valid predictions of HF rehospitalization in 1year follow-up. It can be easily utilized to guide appropriate interventions or care strategies for patients with HF. The closed monitoring and blood test in daily clinics are important for assessing the risk of HF rehospitalization.

Encoding temporal information in deep convolution neural network.

A recent development in deep learning techniques has attracted attention to the decoding and classification of electroencephalogram (EEG) signals. Despite several efforts to utilize different features in EEG signals, a significant research challenge is using time-dependent features in combination with local and global features. Several attempts have been made to remodel the deep learning convolution neural networks (CNNs) to capture time-dependency information. These features are usually either handcrafted features, such as power ratios, or splitting data into smaller-sized windows related to specific properties, such as a peak at 300 ms. However, these approaches partially solve the problem but simultaneously hinder CNNs' capability to learn from unknown information that might be present in the data. Other approaches, like recurrent neural networks, are very suitable for learning time-dependent information from EEG signals in the presence of unrelated sequential data. To solve this, we have proposed an encoding kernel (EnK), a novel time-encoding approach, which uniquely introduces time decomposition information during the vertical convolution operation in CNNs. The encoded information lets CNNs learn time-dependent features in addition to local and global features. We performed extensive experiments on several EEG data sets-physical human-robot collaborations, P300 visual-evoked potentials, motor imagery, movement-related cortical potentials, and the Dataset for Emotion Analysis Using Physiological Signals. The EnK outperforms the state of the art with an up to 6.5% reduction in mean squared error (MSE) and a 9.5% improvement in F1-scores compared to the average for all data sets together compared to base models. These results support our approach and show a high potential to improve the performance of physiological and non-physiological data. Moreover, the EnK can be applied to virtually any deep learning architecture with minimal effort.

Assembly-Induced Dynamic Structural Color in a Host-Guest System for Time-Dependent Anticounterfeiting and Double-Lock Encryption.

Manipulation of periodic micro/nanostructures in polymer film is of great importance for academics and industrial applications in anticounterfeiting. However, with the increasing demand on information security, materials with time-dependent features are urgently required, especially the material where the same information can appear more than once on the time scale. Here, one concise strategy to realize time-dependent anticounterfeiting and "double-lock" information encryption based on a host-guest system is proposed, with one photoresponsive azopolymer as the host and one liquid-crystalline molecule as the guest. The system exhibits a tunable mass transport in pre-designed periodic micro/nanostructures by tailoring the process of cis-to-trans recovery of azo groups and assembly of mesogenic trans-isomers, resulting in a dynamic structural color in film. Taking advantage of this extraordinary feature, time-dependent dynamic anticounterfeiting has been achieved. More importantly, the time of each state's appearance in the whole process can be modulated by changing the host-guest ratio. Combining the manipulatable process of mass transport with the unique decoding method, the stored information in film can be decrypted correctly. This work provides an unprecedented dynamic approach for advanced anticounterfeiting technology with a higher level of security and high-end applications in information encryption.

Convolutional preprocessing Transformer-based fault diagnosis for rectifier-filter circuits in nuclear power plants

Rectifier-filter circuit, as a vital electronic component in nuclear power plants (NPPs), serves the role of converting alternating current into smooth direct current. Affected by intense radiation environmental factors and variable operating profiles, the rectifier-filter circuit is subject to degradation and soft failures, namely, a derated circuit performance but not yet to failure. It is urgent to diagnose the soft failures of the rectifier-filter circuit, so as to ensure the safe operations of NPPs. In this article, an ensemble empirical mode decomposition (EEMD) algorithm is first employed to decompose the three-phase current signals and electrical signals of electronic components. The decomposed signals are fed into a convolutional preprocessing Transformer (CPT) to deeply extract the fault features of the rectifier-filter circuit. Specifically, the multi-directional, multi-scale, and extremely sensitive long-range time-dependent features in the EEMD feature data are extracted by the multi-head attention mechanism of the Transformer. A deep Softmax classifier is, then, devised to reduce the feature dimensionality and identify the soft fault modes of the rectifier-filter circuit. The fault simulation of the rectifier-filter circuit is conducted, and a real case experiment and multiple comparative studies are conducted to validate the effectiveness and diagnosis accuracy of the CPT-based method.

3D clustering of gene expression data from systemic autoinflammatory diseases using self-organizing maps (Clust3D)

Background and objectiveSystemic autoinflammatory diseases (SAIDs) are characterized by widespread inflammation, but for most of them there is a lack of specific biomarkers for accurate diagnosis. Although a number of machine learning algorithms have been used to analyze SAID datasets, aiding in the discovery of novel biomarkers, there is a growing recognition of the importance of SAID timeseries clustering, as it can capture the temporal dynamics of gene expression patterns. MethodologyThis paper proposes a novel clustering methodology to efficiently associate three-dimensional data. The algorithm utilizes competitive learning to create a self-organizing neural network and adjust neuron positions in time-dependent and high dimensional feature space in order to assign them as clustering centers. The quantitative evaluation of the clustering was based on well-known clustering indices. Furthermore, a differential expression analysis and classification pipeline was employed to assess the capability of the proposed methodology to extract more accurate pathway-specific genes from its clusters. For that, a comparative analysis was also conducted against a heuristic timeseries clustering method. ResultsThe proposed methodology achieved better overall clustering indices scores and classification metrics using genes derived from its clusters. Notable cases include a threefold increase in the Calinski-Harabasz clustering index, a twofold improvement in the Davies–Bouldin clustering index and a ∼60% increase in the classification specificity score. ConclusionA novel clustering methodology was developed and applied on several gene expression timeseries datasets from systemic autoinflammatory diseases, and its ability to efficiently produce well separated clusters compared to existing heuristic methods was demonstrated.

Heart Signal Analysis Using Multistage Classification Denoising Model

Cardiovascular disease is a major cause of death worldwide, and the COVID-19 pandemic has only made the situation worse. The purpose of this work is to explore various time-frequency analysis methods that can be used to classify heart sound signals and identify multiple abnormalities in the heart, such as aortic stenosis, mitral stenosis, and mitral valve prolapse. The signal has been modified using three techniques—tunable quality wavelet transform (TQWT), discrete wavelet transform (DWT), and empirical mode decomposition—to detect heart signal abnormality. The proposed model detects heart signal abnormality at two stages, the user end and the clinical end. At the user end, binary classification of signals is performed, and if signals are abnormal then further classification is done at the clinic. The approach starts with signal preprocessing and uses the discrete wavelet transform (DWT) coefficients to train the hybrid model, which consists of one long short-term memory (LSTM) network layer and three convolutional neural network (CNN) layers. This method produced comparable results, with a 98.9% classification accuracy for signals, through the utilization of the CNN and LSTM model. Combining the CNN’s skill in feature extraction with the LSTM’s capacity to record time-dependent features improves the efficacy of the model. Identifying issues early and initiating appropriate medication can alleviate the burden associated with heart valve diseases.

Three-dimensional numerical modeling of the temporal evolution of backward erosion piping

Backward erosion piping (BEP) is a complex degradation mechanism in geotechnical flood protection infrastructure (GFPI) that is still relatively less understood, particularly when considering its time-dependent features. This manuscript presents a novel dual random lattice modeling approach for three-dimensional simulation of BEP, with a focus on its evolution over time. The key novelty of this presented framework is twofold: (1) we propose and incorporate a novel constitutive relationship for computation of time-dependent soil erosion based on the theory of rate processes, and (2) we devise an algorithm for calculation of coupled degradation of the dual lattices for accurate computation of 3-D hydraulic gradients. The constitutive relationship was developed from fundamental granular physics, and brings the potential to provide deeper fundamental physical understanding of the phenomenon. The capabilities of the modeling framework are investigated by comparison with available laboratory experiments which illustrates good agreement in the spatial advancement of piping erosion, pipe progression speeds, as well as the evolution of local gradients. To the best knowledge of the authors, the presented model is the first to be able to capture all of the aforementioned features when simulating BEP.

A Fault Diagnosis Method for Rectifier-Filter Circuit Integrating EEMD Algorithm and Transformer Network

Rectifier-filter circuit, as a critical component of the drive circuit in instrumentation and control systems of nuclear power plants, can convert the 50 Hz AC into the smooth DC. Thus, it plays a vital role in the power control of reactors. However, the weak waveform anomalies of soft faults in the rectifier-filter circuit make fault feature extraction difficult. Therefore, in this article, an ensemble empirical modal decomposition (EEMD) algorithm is employed to decompose the signal mode components in the monitor data of the rectifier-filter circuit. The weak waveform anomalies are indirectly enhanced by the IMF and residual components. Subsequently, the Transformer network is utilized to construct the feature extractor. With the advantage of multi-head attention (MHA) mechanism in the Transformer network, the multi-directional, multi-scale, and highly sensitive long-range time-dependent features in the EEMD feature data are extracted. Then, a deep Softmax classifier is adopted to reduce the dimensionality and diagnose the soft faults of the rectifier-filter circuit. Finally, a fault simulation model of the rectifier-filter circuit is constructed and the condition monitor data are collected. The effectiveness and diagnosis accuracy of the proposed method are verified by a real case experiment and some comparative methods.

Neural network-based surrogate modeling and optimization of a multigeneration system

Multi-Objective Optimization (MOO) poses a computational challenge, particularly when applied to physics-based models. As a result, only up to three objectives are typically involved in simulation-based optimization. To go beyond this number, Surrogate Models (SMs) need to replace such high-fidelity models. In this exploratory study, the objectives are to perform comprehensive regression surrogate modeling and to conduct MOO for a Multi-Generation System (MGS). The most suitable SM was chosen among four neural-network models: Artificial Neural Network (ANN), Convolutional Neural Network (CNN), Long-Short Term Memory (LSTM), and an ensemble model developed through brute-force search using the three aforementioned models. The final model was found to be superior to others, achieving R2 values ranging from 0.9830 to 0.9999. Next, an optimization problem with six conflicting objectives was defined and performed at four distinct values of Direct Normal Irradiation (DNI), a time-dependent feature. This aimed to provide multi-criteria decision-making information based on atmospheric transparency. As a result, new understandings were gained: (I) exergy efficiency, production cost, and freshwater production rate were found to be highly influenced by DNI, and (II) the critical range of operation was observed within the DNI interval of 100 to 400 W/m2. Furthermore, we compared the result of the six-objective optimization with that of the bi-objective optimization obtained in our simulation-based study and found that all objectives showed improvements ranging from 1.9% to 12.7%. Finally, based on the findings obtained in the present study, some practical recommendations were put forward for applying the proposed methodology to similar MGSs.

Improving prediction of 12-months suicidal attempts in bipolar disorder: a machine learning study

IntroductionBipolar disorder (BD) is a recurrent disorder, causing functional impairment and raised mortality, particularly due to suicide. However, the difficulty in predicting suicidal behaviors relies in the lack of clear biomarkers.Machine learning (ML) has emerged as a promising tool to enhance suicidal prediction. However, most ML studies focused on lifetime attempts, without having a predictive time window, and did not employ time-dependent variables. Moreover, most studies lie on cross-sectional databases, without including more than one time-point.Objectives First, we aimed to predict 12-months suicide attempts in a naturalistic sample of BD patients, using clinical and demographic data.Second, we aimed to improve the prediction by including information from intermediate visits (1, 3, and 6 months), mimicking more closely the clinician’s way of thinking and the multiple observations a patient receives.MethodsA sample of 163 BD patients (53% females, mean age 44.7, SD 15.3) were recruited.Based on EHR, 56 clinical and demographic features were extracted, including hospitalizations, suicidal behaviors lifetime and in the last 12 months, along with comorbidity, family history, work, and therapies. Patients were followed up for 12 months.Support Vector Machine (SVM) was used to differentiate subjects who attempted suicide versus those who did not in a 12-month time window, within a repeated nested Cross-Validation. The SVM was optimized weighting the hyperplane for uneven group sizes.Then, we repeated the analysis including information from intermediate visits (1, 3, 6 months after the first contact). For each visit, we created a composite score based on current therapy, new admissions, and ER presentations. To avoid circularity, all the information (ER, admission etc.) related to a suicide attempt were not included.ResultsDuring the 12-months follow-up, 9.8% of patients attempted suicide. The results from the 12-months suicide prediction model obtained an Area Under the Curve of 0.71(with a Balanced Accuracy (BAC) of 68%).After incorporating the composite scores based on intermediate visits in the model, the prediction raised to an Area Under the Curve of 0.78 (BAC 73%), suggesting that including intermediate visits is a valid method to improve prediction.The features that contributed the most to the prediction were the composite score at 6-month visit, lifetime number of suicide attempts, suicide attempts in the last 12 months, substance of abuse (other than cannabis), and antipsychotics.ConclusionsML proved a good prediction accuracy for suicide in a 12-months time window, and the prediction was improved by including data from intermediate visits. The model showed the importance of time-dependent features, such as attempts in the last 12 months. Our analysis might help in identifying early clinical risk factors and underlies the importance of multiple evaluations in populations at risk.Disclosure of InterestNone Declared

Dynamically coupled kinetic chemistry in brown dwarf atmospheres – II. Cloud and chemistry connections in directly imaged sub-Jupiter exoplanets

ABSTRACT With JWST slated to gain high-fidelity time-dependent data on brown dwarf atmospheres, it is highly anticipated to do the same for directly imaged, sub-Jupiter exoplanets. With this new capability, the need for a full three-dimensional (3D) understanding to explain spectral features and their time dependence is becoming a vital aspect for consideration. To examine the atmospheric properties of directly imaged sub-Jupiter exoplanets, we use the 3D Exo-FMS general circulation model to simulate a metal-enhanced generic young sub-Jupiter object. We couple Exo-FMS to a kinetic chemistry scheme, a tracer-based cloud formation scheme and a spectral radiative-transfer model to take into account the chemical and cloud feedback on the atmospheric thermochemical and dynamical properties. Our results show a highly complex feedback between clouds and chemistry on to the 3D temperature structure of the atmosphere, bringing about latitudinal differences and inducing time-dependent stormy features at photospheric pressures. This suggests a strong connection and feedback between the spatial cloud coverage and chemical composition of the atmosphere, with the temperature changes and dynamical motions induced by cloud opacity and triggered convection feedback driving chemical species behaviour. In addition, we also produce synthetic latitude-dependent and time-dependent spectra of our model to investigate atmospheric variability and periodicity in commonly used photometric bands. Overall, our efforts put the included physics in 3D simulations of exoplanets on par with contemporary 1D radiative-convective equilibrium modelling.

A unified deep learning framework for water quality prediction based on time-frequency feature extraction and data feature enhancement

Deep learning methods exhibited significant advantages in mapping highly nonlinear relationships with acceptable computational speed, and have been widely used to predict water quality. However, various model selection and construction methods resulted in differences in prediction accuracy and performance. Hence, a unified deep learning framework for water quality prediction was established in the paper, including data processing module, feature enhancement module, and data prediction module. In the established model, the data processing module based on wavelet transform method was applied to decomposing complex nonlinear meteorology, hydrology, and water quality data into multiple frequency domain signals for extracting self characteristics of data cyclic and fluctuations. The feature enhancement module based on Informer Encoder was used to enhance feature encoding of time series data in different frequency domains to discover global time dependent features of variables. Finally, the data prediction module based on the stacked bidirectional long and short term memory network (SBiLSTM) method was employed to strengthen the local correlation of feature sequences and predict the water quality. The established model framework was applied in Lijiang River in Guilin, China. The maximum relative errors between the predicted and observed values for dissolved oxygen (DO), chemical oxygen demand (CODMn) were 12.4% and 20.7%, suggesting a satisfactory prediction performance of the established model. The validation results showed that the established model was superior to all other models in terms of prediction accuracy with RMSE values 0.329, 0.121, MAE values 0.217, 0.057, SMAPE values 0.022, 0.063 for DO and CODMn, respectively. Ablation tests confirmed the necessity and rationality of each module for the established model framework. The established method provided a unified deep learning framework for water quality prediction.

Short-term influence of polytetrafluoroethylene micro/nano-plastics on the inhibition of copper and/or ciprofloxacin on the nitrifying sludge activities based on concentration addition and independent action models

Short-term influence of polytetrafluoroethylene micro/nano-plastics (PTFE-MPs/NPs) on the inhibition of copper (Cu2+) and/or ciprofloxacin (CIP) on the nitrifying sludge activities was explored based on concentration addition (CA) and independent action (IA) models. The half maximal inhibitory concentration (IC50) of Cu2+, CIP, PTFE-MPs (3 μm), and PTFE-NPs (800 nm) on the specific ammonium oxidation rate (SAOR) of nitrifying sludge was 64.57, 51.29, 102.33 and 93.33 mg L−1, respectively, while those on the specific nitrite oxidation rate (SNOR) of nitrifying sludge were 77.62, 32.36, 104.70 and 97.72 mg L−1, respectively. Among the five binary mixtures and two ternary mixtures composed by Cu2+, CIP, and/or PTFE-MPs/NPs, it was found that the two joint inhibitory actions from ternary mixtures on the SAOR and SNOR of the sludge showed time-dependent characteristics by analyzing of CA and IA models, while the five combined inhibitory effects from different binary mixtures did not all have time-dependent features. The two joint inhibition actions from diverse ternary mixtures on the SAOR at the exposure time of 60 min and on the SNOR at 90 min showed always concentration-dependent features, while the combined inhibitions with concentration-dependent characteristics had never been observed in the binary Cu2+ and PTFE-NPs mixtures at different exposure time. The Cu2+, CIP, and PTFE-MPs mixtures (or Cu2+, CIP, and PTFE-NPs mixtures) had synergistic actions on the SAOR at 90 min and antagonistic effects on the SNOR at 60 min based on CA and IA models, and these combined inhibitions did not exhibit concentration-dependent characteristics. In contrast, the joint inhibitory effects (on the SAOR and SNOR) with concentration-dependent features were found in the binary mixtures of CIP and PTFE-MPs at different exposure time, and the join inhibition changed from synergism to antagonism as the increasing concentration of mixed CIP and PTFE-MPs. This study provides novel perspectives for understanding the combined influence of plastic particles with different sizes, antibiotics, and heavy metals on the biological wastewater treatment process.

Theoretical Study of the Two-Dimensional Vibrational Sum Frequency Generation Spectroscopy of the Air-Water Interface at Varying Temperature and Its Connections to the Interfacial Structure and Dynamics.

We performed a theoretical study of the temperature variation of two-dimensional vibrational sum frequency generation (2D-VSFG) spectra of the OH stretch modes at air-water interfaces in the mid-IR region. The calculations are performed at four different temperatures from 250 to 325 K by using a combination of techniques involving response function formalism of nonlinear spectroscopy, electronic structure calculations, and molecular dynamics simulations. Also, the calculations are performed for isotopically dilute solutions so that the intra- and intermolecular coupling between the vibrational modes of interest can be ignored. We have established the connections of temperature variation of various frequency- and time-dependent features of the calculated spectra to the changes in the underlying structure and dynamics of the interfaces. The results reveal that interfacial water is dynamically more heterogeneous than bulk water, with three dominant dynamical processes exhibiting their corresponding time-dependent features in the 2D-VSFG spectrum. These are the spectral diffusion of hydrogen-bonded OH groups at the interface, conversion of an initially hydrogen-bonded OH group to a dangling OH which is a stable state for surface water, unlike the bulk water, and the third one, which involves the conversion of an initially free or dangling OH group to its hydrogen-bonded state at the interface. The temporal appearance of the cross peaks corresponding to interconversion of the hydrogen-bonded state to the dangling state or vice versa of an interfacial OH group is found to take place at a slower rate than the dynamics of spectral diffusion of hydrogen-bonded molecules at the interface, which, in turn, is slower than the corresponding spectral diffusion of bulk water molecules. The temperature variation of these dynamic processes can be linked to the decay of appropriate hydrogen-bond and non-hydrogen-bond time correlation functions of interfacial water molecules for the different air-water systems studied in this work.

Optimizing platelet transfusion through a personalized deep learning risk assessment system for demand management

AbstractPlatelet demand management (PDM) is a resource-consuming task for physicians and transfusion managers of large hospitals. Inpatient numbers and institutional standards play significant roles in PDM. However, reliance on these factors alone commonly results in platelet shortages. Using data from multiple sources, we developed, validated, tested, and implemented a patient-specific approach to support PDM that uses a deep learning–based risk score to forecast platelet transfusions for each hospitalized patient in the next 24 hours. The models were developed using retrospective electronic health record data of 34 809 patients treated between 2017 and 2022. Static and time-dependent features included demographics, diagnoses, procedures, blood counts, past transfusions, hematotoxic medications, and hospitalization duration. Using an expanding window approach, we created a training and live-prediction pipeline with a 30-day input and 24-hour forecast. Hyperparameter tuning determined the best validation area under the precision-recall curve (AUC-PR) score for Long Short-Term Memory deep learning models, which were then tested on independent data sets from the same hospital. The model tailored for hematology and oncology patients exhibited the best performance (AUC-PR, 0.84; area under the receiver operating characteristic curve [ROC-AUC], 0.98), followed by a multispecialty model covering all other patients (AUC-PR, 0.73). The model specific to cardiothoracic surgery had the lowest performance (AUC-PR, 0.42), likely because of unexpected intrasurgery bleedings. To our knowledge, this is the first deep learning–based platelet transfusion predictor enabling individualized 24-hour risk assessments at high AUC-PR. Implemented as a decision-support system, deep-learning forecasts might improve patient care by detecting platelet demand earlier and preventing critical transfusion shortages.

State-of-health estimation of lithium-ion batteries using a novel dual-stage attention mechanism based recurrent neural network

Accurate estimation of the state of health (SOH) of lithium-ion batteries is an important guarantee to ensure safe and reliable operation of lithium-ion battery systems. However, the complex aging mechanism inside the battery makes it difficult to measure the battery SOH directly. In this paper, a SOH estimation method based on a novel dual-stage attention-based recurrent neural network (DARNN) and health feature (HF) extraction from time varying charging process is proposed. Firstly, the constant current charging time, the maximum temperature time, the isochronous voltage difference, and the isochronous current were extracted as lithium-ion battery HFs, and their correlations with SOH are verified by spearman correlation coefficient. Secondly, the DARNN is proposed to capture the time-dependent and temporal features of the input sequence and to accurately predict SOH. Finally, the proposed estimation method is validated on the NASA battery dataset. The results show that the method can accurately estimate SOH for lithium-ion batteries. The mean square error and the mean absolute percentage error of the method are <0.5 %.

Malfunction diagnosis of main station of power metering system using LSTM-ResNet with SMOTE method

The power metering system is an important part of the smart grid for data acquisition and analysis. The fault state of the main station directly affects the stable and safe operation of the power metering system. Hinged on the real-world data supplied by the monitoring platform of the Metrology Center of Guangdong Power Grid Co., Ltd., we present a novel malfunction diagnosis method for the main station of the power metering system. The proposed method utilizes the synthetic mi-nority over-sampling technique (SMOTE) and designs a combined model of long short-term memory (LSTM) network and ResNet. SMOTE solves the sample imbalance problem. Furthermore, the combined LSTM-ResNet model employs LSTM to extract the time-dependent signal feature and exploits ResNet to optimize data flow. Consequently, the proposed LSTM-ResNet model improves training efficiency and malfunction diagnosis accuracy. The proposed diagnosis mthod is verifird on the real-world data, which proves the proposed method’s surpass traditional methods. A specific analysis of results and the practical application of the proposed method is also elaborated.

Application of Kalman Filtering with Bayesian formulation in adaptive sampling

An extensive amount of research is emphasized on survey designs and estimation procedures related to rare and clustered characteristics of a population. Adaptive Sampling design is the most applicable probabilistic technique to estimate the mean or total of the variable of interest, bearing rarity and clustered characteristics. Since rarity is regarded as a time-dependent feature, such surveys need to be organized constantly over time. No studies so far have investigated the effect of time in the estimation context of Adaptive Sampling. This research therefore captures the need to synthesize this periodic information when conducting a survey using Adaptive Sampling design. A recursive process is employed here that improves the estimate of the population parameter from a practical perspective. “Kalman Filtering” is a well known recursive procedure to use past data. Later, statisticians were able to use that Kalman Filtering technique with the Bayesian formulation. This Bayesian approach is proposed to employ here to improve the estimation in the context of Adaptive Sampling design, utilizing the past data. A simulation study is carried out and it is concluded that the suggested approach substantially improves the estimation accuracy.

On the relative importance of shocks and self-gravity in modifying tidal disruption event debris streams

ABSTRACT In a tidal disruption event (TDE), a star is destroyed by the gravitational field of a supermassive black hole (SMBH) to produce a stream of debris, some of which accretes onto the SMBH and creates a luminous flare. The distribution of mass along the stream has a direct impact on the accretion rate, and thus modelling the time-dependent evolution of this distribution provides insight into the relevant physical processes that drive the observable properties of TDEs. Analytic models that only account for the ballistic evolution of the debris do not capture salient and time-dependent features of the mass distribution, suggesting that fluid dynamical effects significantly modify the debris dynamics. Previous investigations have claimed that shocks are primarily responsible for these modifications, but here we show – with high-resolution hydrodynamical simulations – that self-gravity is the dominant physical mechanism responsible for the anomalous (i.e. not predicted by ballistic models) debris stream features and its time dependence. These high-resolution simulations also show that there is a specific length-scale on which self-gravity modifies the debris mass distribution, and as such there is enhanced power in specific Fourier modes. Our results have implications for the stability of the debris stream under the influence of self-gravity, particularly at late times and the corresponding observational signatures of TDEs.