Search

In this study, we introduce a novel approach to speech enhancement through the design of a complex temporal convolutional network (Complex-TCN). This model leverages the power of complex networks, enabling the simultaneous capture of both magnitude and phase information inherent in speech signals. By employing a temporal convolutional network, the Complex-TCN excels at extracting contextual information within the time domain of speech. Our findings underscore the substantial performance improvements achieved through the synergistic use of the temporal convolutional network and the incorporation of complex representations. [ABSTRACT FROM AUTHOR]

Accurate and reliable PV power probabilistic-forecasting results can help grid operators and market participants better understand and cope with PV energy volatility and uncertainty and improve the efficiency of energy dispatch and operation, which plays an important role in application scenarios such as power market trading, risk management, and grid scheduling. In this paper, an innovative deep learning quantile regression ultra-short-term PV power-forecasting method is proposed. This method employs a two-branch deep learning architecture to forecast the conditional quantile of PV power; one branch is a QR-based stacked conventional convolutional neural network (QR_CNN), and the other is a QR-based temporal convolutional network (QR_TCN). The stacked CNN is used to focus on learning short-term local dependencies in PV power sequences, and the TCN is used to learn long-term temporal constraints between multi-feature data. These two branches extract different features from input data with different prior knowledge. By jointly training the two branches, the model is able to learn the probability distribution of PV power and obtain discrete conditional quantile forecasts of PV power in the ultra-short term. Then, based on these conditional quantile forecasts, a kernel density estimation method is used to estimate the PV power probability density function. The proposed method innovatively employs two ways of a priori knowledge injection: constructing a differential sequence of historical power as an input feature to provide more information about the ultrashort-term dynamics of the PV power and, at the same time, dividing it, together with all the other features, into two sets of inputs that contain different a priori features according to the demand of the forecasting task; and the dual-branching model architecture is designed to deeply match the data of the two sets of input features to the corresponding branching model computational mechanisms. The two a priori knowledge injection methods provide more effective features for the model and improve the forecasting performance and understandability of the model. The performance of the proposed model in point forecasting, interval forecasting, and probabilistic forecasting is comprehensively evaluated through the case of a real PV plant. The experimental results show that the proposed model performs well on the task of ultra-short-term PV power probabilistic forecasting and outperforms other state-of-the-art deep learning models in the field combined with QR. The proposed method in this paper can provide technical support for application scenarios such as energy scheduling, market trading, and risk management on the ultra-short-term time scale of the power system. [ABSTRACT FROM AUTHOR]

Copyright of Electric Power Automation Equipment / Dianli Zidonghua Shebei is the property of Electric Power Automation Equipment Press and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Due to the frequent occurrence of extreme weather in recent years, accurate runoff prediction is crucial for the rational planning and management of water resources. Addressing the high uncertainty and multiple influencing factors in runoff prediction, this paper proposes a runoff prediction method driven by multi-source data. Based on multivariate observed data of runoff, water level, temperature, and precipitation, a Time2Vec-TCN-Transformer model is proposed for runoff prediction research and compared with LSTM, TCN, and TCN-Transformer models. The results show that the Time2Vec-TCN-Transformer model outperforms other models in metrics including MAE, RRMSE, MAPE, and NSE, demonstrating higher prediction accuracy and reliability. By effectively combining Time2Vec, TCN, and Transformer, the proposed model improves the MAPE for forecasting 1–4 days in the future by approximately 7% compared to the traditional LSTM model and 4% compared to the standalone TCN model, while maintaining NSE consistently between 0.9 and 1. This model can better capture the periodicity, long-term scale information, and relationships among multiple variables of runoff data, providing reliable predictive support for flood forecasting and water resources management. [ABSTRACT FROM AUTHOR]

Climate change is causing water levels to rise, leading to detrimental effects like tidal flooding in coastal areas. Surabaya, the capital of East Java Province in Indonesia, is particularly vulnerable due to its low-lying location. According to the Meteorological, Climatological, and Geophysical Agency (BMKG), tidal flooding occurs annually in Surabaya as a result of rising water levels, highlighting the urgent need for water level forecasting models to mitigate these impacts. In this study, we employ the Temporal Convolutional Network (TCN) machine learning model for water level forecasting using data from a sea level station monitoring facility in Surabaya. We divided the training data into three scenarios: 3, 6, and 8 months to train TCN models for 14-day forecasts. The 8-month training scenario yielded the best results. Subsequently, we used the 8-month training data to forecast 1, 3, 7, and 14 days using TCN, Transformers, and the Recurrent Neural Network (RNN) models. TCN consistently outperformed other models, particularly excelling in 1-day forecasting with coefficient of determination () and RMSE values of 0.9950 and 0.0487, respectively.

The power industry has been dedicated to applying virtual reality (VR) technology to build training systems in virtual environments, enabling personnel to complete skill training in real simulated environments while ensuring their safety. Conventional action scoring systems struggle to provide accurate scores for fine movements. Accurate scoring of fine movements can help workers identify their shortcomings during power operations, thus improving learning efficiency. This is of great significance for training on virtual environment-based power operation. This paper proposes a power operation-orientated VR action evaluation method based on the Gaussian distance-improved dynamic time warping (DTW) algorithm and the temporal convolutional network (TCN) model. First, the adaptive adapter is used to extract one-dimensional features from the three-dimensional data of the data gloves. Then, based on the TCN model, action data with significant discrepancies are filtered out. Finally, the obtained data are input into the Gaussian distance-improved DTW algorithm, where the path size is calculated. Corresponding scoring criteria are established on the basis of the path size to evaluate the actions. The results demonstrate that the VR action evaluation method based on the Gaussian distance-improved DTW algorithm and the TCN model significantly improves the accuracy of evaluating fine movements compared to traditional evaluation algorithms.

Artificial Intelligence (AI) techniques leverage the justice system in terms of effectiveness and efficiency. AI-empowered multi-label crime classification can facilitate the precise and expedient categorization of various legal documents. Multi-label classification within the justice system has an indispensable role in achieving accurate legal categorization that invigorates case analysis, optimizes resource distribution and refines the contours of legal processes. To support this critical function, this paper proposes a temporal convolutional network (TCN)-bidirectional encoder representations from transformers (BERT)-coupled model for multi-label crime classification. The proposed method fuses the temporal formation and semantic information in the model to obtain a high-quality result. The experimental results show that the proposed method achieved the best accuracy in comparison to existing methods on a public dataset. [ABSTRACT FROM AUTHOR]

Photovoltaic (PV) power forecasting plays a crucial role in optimizing renewable energy integration into the grid, necessitating accurate predictions to mitigate the inherent variability of solar energy generation. We propose a novel forecasting model that combines improved variational mode decomposition (IVMD) with the temporal convolutional network-gated recurrent unit (TCN-GRU) architecture, enriched with a multi-head attention mechanism. By focusing on four key environmental factors influencing PV output, the proposed IVMD-TCN-GRU framework targets a significant research gap in renewable energy forecasting methodologies. Initially, leveraging the sparrow search algorithm (SSA), we optimize the parameters of VMD, including the mode component K-value and penalty factor, based on the minimum envelope entropy principle. The optimized VMD then decomposes PV power, while the TCN-GRU model harnesses TCN's proficiency in learning local temporal features and GRU's capability in rapidly modeling sequence data, while leveraging multi-head attention to better utilize the global correlation information within sequence data. Through this design, the model adeptly captures the correlations within time series data, demonstrating superior performance in prediction tasks. Subsequently, the SSA is employed to optimize GRU parameters, and the decomposed PV power mode components and environmental feature attributes are inputted into the TCN-GRU neural network. This facilitates dynamic temporal modeling of multivariate feature sequences. Finally, the predicted values of each component are summed to realize PV power forecasting. Validation using real data from a PV station corroborates that the novel model demonstrates a substantial reduction in RMSE and MAE of up to 55.1% and 54.5%, respectively, particularly evident in instances of pronounced photovoltaic power fluctuations during inclement weather conditions. The proposed method exhibits marked improvements in accuracy compared to traditional PV power prediction methods, underscoring its significance in enhancing forecasting precision and ensuring the secure scheduling and stable operation of power systems. [ABSTRACT FROM AUTHOR]

Attention Deficit Hyperactivity Disorder (ADHD) is a prevalent neurodevelopmental disorder that significantly affects children and adults worldwide, characterized by persistent inattention, hyperactivity, and impulsivity. Current research in this field faces challenges, particularly in accurate diagnosis and effective treatment strategies. The analysis of motor information, enriched by artificial intelligence methodologies, plays a vital role in deepening our understanding and improving the management of ADHD. The integration of AI techniques, such as machine learning and data analysis, into the study of ADHD-relatedmotor behaviors, allows for amore nuanced understanding of the disorder. This approach facilitates the identification of patterns and anomalies in motor activity that are often characteristic of ADHD, thereby contributing to more precise diagnostics and tailored treatment strategies. Our approach focuses on utilizing AI techniques to deeply analyze patients' motor information and cognitive processes, aiming to improve ADHD diagnosis and treatment strategies. On the ADHD dataset, the model significantly improved accuracy to 98.21% and recall to 93.86%, especially excelling in EEG data processing with accuracy and recall rates of 96.62 and 95.21%, respectively, demonstrating precise capturing of ADHD characteristic behaviors and physiological responses. These results not only reveal the great potential of ourmodel in improving ADHD diagnostic accuracy and developing personalized treatment plans, but also open up new research perspectives for understanding the complex neurological logic of ADHD. In addition, our study not only suggests innovative perspectives and approaches for ADHD treatment, but also provides a solid foundation for future research exploring similar complex neurological disorders, providing valuable data and insights. This is scientifically important for improving treatment outcomes and patients' quality of life, and points the way for future-oriented medical research and clinical practice. [ABSTRACT FROM AUTHOR]

Automotive radar is one of the key sensors for intelligent driving. Radar image sequences contain abundant spatial and temporal information, enabling target classification. For existing radar spatiotemporal classifiers, multi-view radar images are usually employed to enhance the information of the target and 3D convolution is employed for spatiotemporal feature extraction. These models consume significant hardware resources and are not applicable to real-time applications. In this paper, RadarTCN, a novel lightweight network, is proposed that achieves high-accuracy online target classification using single-view radar image sequences only. In RadarTCN, 2D convolution and 3D-TCN are employed to extract spatiotemporal features sequentially. To reduce data dimensionality and computational complexity, a multi-layer max pooling down-sampling method is designed in a 2D convolution module. Meanwhile, the 3D-TCN module is improved through residual pruning and causal convolution is introduced for leveraging the performance of online target classification. The experimental results demonstrate that RadarTCN can achieve high-precision online target recognition for both range-angle and range-Doppler map sequences. Compared to the reference models on the CARRADA dataset, RadarTCN exhibits better classification performance, with fewer parameters and lower computational complexity. [ABSTRACT FROM AUTHOR]

The widespread and substantial effect of the global financial crisis in history underlines the importance of forecasting financial crisis effectively. In this paper, we propose temporal convolutional network (TCN), which based on a convolutional neural network, to construct an early warning system for financial crises. The proposed TCN is compared with logit model and other deep learning models. The Shapley value decomposition is calculated for the interpretability of the early warning system. Experimental results show that the proposed TCN outperforms other models, and the stock price and the real GDP growth have the largest contributions in the crises prediction. [ABSTRACT FROM AUTHOR]

Accurate prediction of the Remaining Useful Life (RUL) of lithium-ion batteries is crucial for reducing battery usage risks and ensuring the safe operation of systems. Addressing the impact of noise and capacity regeneration-induced nonlinear features on RUL prediction accuracy, this paper proposes a predictive model based on Complete Ensemble Empirical Mode Decomposition with Adaptive Noise (CEEMDAN) data preprocessing and IHSSA-LSTM-TCN. Firstly, CEEMDAN is used to decompose lithium-ion battery capacity data into high-frequency and low-frequency components. Subsequently, for the high-frequency component, a Temporal Convolutional Network (TCN) prediction model is employed. For the low-frequency component, an Improved Sparrow Search Algorithm (IHSSA) is utilized, which incorporates iterative chaotic mapping and a variable spiral coefficient to optimize the hyperparameters of Long Short-Term Memory (LSTM). The IHSSA-LSTM prediction model is obtained and used for prediction. Finally, the predicted values of the sub-models are combined to obtain the final RUL result. The proposed model is validated using the publicly available NASA dataset and CALCE dataset. The results demonstrate that this model outperforms other models, indicating good predictive performance and robustness. [ABSTRACT FROM AUTHOR]

At present, the prediction accuracy of a single model is difficult to remain optimal with power fluctuation. To improve the stability of grid connected system operation and energy saving dispatching of power grid, this study proposes a dynamic learning combination short-term power prediction method based on k-sums hierarchical clustering. The weather types are divided into sunny day A1, cloudy day A2, and rainy day B through segmentation clustering using k-sums algorithm. The TCN (Temporal Convolutional Network) is used to extract the time series characteristics of data, and the GRU(Gate Recurrent Unit) structure of the fusion extraction time series characteristics module is established with GRU to achieve the effect of being sensitive to the time series characteristics. After dynamically combining the improved GRU structure with the SVM(Support Vector Machine), the Elastic Net algorithm is adopted to output the best weight value to obtain the final prediction value. The power data of photovoltaic power generation and corresponding meteorological data of a region in Jiangsu are used to verify the proposed method. The results show that the MAE(Mean Absolute Error) of the dynamic combination learning model is 1.888, and the RMSE(Root Mean Squared Error) is 2.403. [ABSTRACT FROM AUTHOR]

BackgroundAcute upper respiratory infection (AURI) is a significant disease affecting all age groups worldwide. The differences in the impacts of different temperature change indicators, such as diurnal temperature range (DTR), temperature variation (TV), and temperature change between neighboring days (TCN), on AURI morbidity, are not clear.MethodsWe collected data on 87,186 AURI patients during 2014–2019 in Zhengzhou. Distributed lag non-linear model was adopted to examine the effects of different temperature change indicators on AURI. We calculated and compared the attributable fractions (AF) of AURI morbidity caused by various indicators. We used stratified analysis to investigate the modification effects of season and gender.ResultsWith the increase in DTR and TV, the risk of AURI tended to increase; the corresponding AF values (95% eCI) higher than the references (5% position of the DTR or TV distribution) were 24.26% (15.46%, 32.05%), 23.10% (15.59%, 29.20%), and 19.24% (13.90%, 24.63%) for DTR, TV0 − 1, and TV0 − 7, respectively. The harmful effects of TCN on AURI mainly occurred when the temperature dropped (TCN < 0), and the AF value of TCN below the reference (0°C) was 3.42% (1.60%, 5.14%). The harm of DTR and TV were statistically significant in spring, autumn and winter, but not in summer, while the harm of TCN mainly occurred in winter. Three indicators have statistically significant effects on both males and females.ConclusionsHigh DTR and TV may induce AURI morbidity, while the harm of TCN occurs when the temperature drops. The impacts of DTR and TV on AURI are higher than that of TCN, and the impact of few-day TV is higher than that of multi-day TV. The adverse effects of DTR and TV are significant except in summer, while the hazards of TCN mainly occur in winter.

Objective Incomplete landslide timing information can result in inaccuracies in the temporal relationship between landslides and rainfall, consequently affecting the precision of a critical rainfall threshold model. Methods To address this issue, this study focuses on rainfall-induced landslides in the Wanzhou District of Chongqing from 1995 to 2015. The Henghe Township, lacking historical landslide data, serves as the verification area. We proposed a prediction model for the daily temporal probability of landslide occurrence on a certain day based on long short-term memory (LSTM) and a temporal convolutional network (TCN). This method was used to reconstruct the temporal information of rainfall-induced landslide events by simulating the nonlinear relationship between the duration of landslides and rainfall. After the reconstruction of temporal information, the landslide events were verified and selected and subsequently applied to a reasonable division of the E-D effective rainfall threshold curve to establish the landslide meteorological warning model. Results The results showed that the average temporal probability of rainfall-induced landslides predicted by the proposed method reached 90.33%, which was higher than that of the ANN (71.17%), LSTM (72.75%), and TCN (86.91%) models. Using temporal probabilities exceeding a 90% threshold, 18 data points, including 42 landslides in the verification area, are expanded to 201. Compared with using solely historical landslide events, the meteorological warning model based on expanded temporal information provides a more reasonable warning classification, and the effective warning rate at the severe warning level is increased by 42.86%. Conclusion This method can compensate for the shortage of landslide time information in field investigations and provide data support for early meteorological warning systems for rainfall-induced landslides, thus improving the accuracy of early meteorological warning systems.

Abstract Due to the uncertainty of weather conditions and the nonlinearity of high-dimensional data, as well as the need for a continuous and stable power supply to the power system, traditional regression analysis and time series forecasting methods are no longer able to meet the high accuracy requirements of today's PV power forecasting. To significantly improve the prediction accuracy of short-term PV output power, this paper proposes a short-term PV power forecasting method based on a hybrid model of temporal convolutional networks and gated recurrent units with an efficient channel attention network (TCN-ECANet-GRU) using the generated data of an Australian PV power station as the research object. First, temporal convolutional networks (TCNs) are used as spatial feature extraction layers, and an efficient channel attention network (ECANet) is embedded to enhance the feature capture capability of the convolutional network. Then, the GRU is used to extract the timing information for the final prediction. Finally, based on the experimental validation, the TCN-ECANet-GRU method generally outperformed the other baseline models in all four seasons of the year according to three performance assessment metrics: the normalized root mean square error (RMSE), normalized mean absolute error (MAE) and coefficient of determination (R2). The best RMSE, MAE and R 2 reached 0.0195, 0.0128 and 99.72%, respectively, with maximum improvements of 11.32%, 8.57% and 0.38%, respectively, over those of the suboptimal model. Therefore, the model proposed in this paper is effective at improving prediction accuracy. Using the proposed method, this paper concludes with multistep predictions of 3, 6, and 9 steps, which also indicates that the proposed method significantly outperforms the other models.

Due to the uncertainty of weather conditions and the nonlinearity of high-dimensional data, as well as the need for a continuous and stable power supply to the power system, traditional regression analysis and time series forecasting methods are no longer able to meet the high accuracy requirements of today's PV power forecasting. To significantly improve the prediction accuracy of short-term PV output power, this paper proposes a short-term PV power forecasting method based on a hybrid model of temporal convolutional networks and gated recurrent units with an efficient channel attention network (TCN-ECANet-GRU) using the generated data of an Australian PV power station as the research object. First, temporal convolutional networks (TCNs) are used as spatial feature extraction layers, and an efficient channel attention network (ECANet) is embedded to enhance the feature capture capability of the convolutional network. Then, the GRU is used to extract the timing information for the final prediction. Finally, based on the experimental validation, the TCN-ECANet-GRU method generally outperformed the other baseline models in all four seasons of the year according to three performance assessment metrics: the normalized root mean square error (RMSE), normalized mean absolute error (MAE) and coefficient of determination (R2). The best RMSE, MAE and R2 reached 0.0195, 0.0128 and 99.72%, respectively, with maximum improvements of 11.32%, 8.57% and 0.38%, respectively, over those of the suboptimal model. Therefore, the model proposed in this paper is effective at improving prediction accuracy. Using the proposed method, this paper concludes with multistep predictions of 3, 6, and 9 steps, which also indicates that the proposed method significantly outperforms the other models. [ABSTRACT FROM AUTHOR]

During the hot rolling process, the work rolls suffer severe wear, resulting in a relatively short lifespan. Severe roll wear can adversely affect the strip shape while introducing roll wear into the crown calculation model can enhance the model accuracy. Therefore, it is crucial to quantify roll wear during the rolling process. Roll wear is a nonlinear time series and the accuracy of the existing work roll wear mechanistic models is not high. In this paper, a novel prediction model for work roll wear based on TCN-LSTM-Attention is developed. TCN utilizes convolutional structures of local and global information to extract data features, while LSTM focuses on capturing long-term and more complex sequence patterns, effectively handling nonlinear characteristics in data. With the incorporation of attention mechanisms, the model becomes more adept at effectively capturing relationships among different segments within the input sequence, which significantly improves predictive performance and reduces the risk of overfitting. Firstly, outlier cleaning and feature selection are performed using Boruta to construct the data set. Then, the predicted results of the proposed model are compared with the existing time series prediction models. The results indicate that the TCN-LSTM-Attention has the highest prediction accuracy, with an R2 of 0.989 and an RMSE of 0.0082 μm. Finally, the predicted results of work roll wear are combined with the mechanism to correct the strip crown pre-calculation model, which significantly improves the calculation accuracy. [ABSTRACT FROM AUTHOR]

Simple Summary: This study addresses the critical need for accurate prediction of key environmental factors—temperature, humidity, ammonia, and hydrogen sulfide—in pig houses, essential for pigs' growth and health. Traditional methods face challenges in predictive accuracy and stability. We introduce an innovative OTDBO–TCN–GRU model, a hybrid framework combining the dung beetle algorithm, temporal convolutional network, and gated recurrent unit, enhanced by the Osprey optimization algorithm (OOA). This model synergistically merges DBO's optimization power, TCN's long-term dependency handling, and GRU's proficiency in nonlinear sequence management, offering improved global detection capabilities. The OTDBO–TCN–GRU model showcases superior accuracy in environmental prediction, evident from its mean absolute error (MAE) of 0.0474, mean squared error (MSE) of 0.0039, and correlation coefficient of 0.9871. It significantly surpasses the traditional DBO–TCN–GRU and OOA models, reducing MAE and MSE by 37.2% and 66.7%, and 48.7% and 74.2%, respectively. Moreover, it outperforms mainstream models like GRU, LSTM, and XGBoost in terms of accuracy. This model significantly improves the forecasting of environmental conditions within pig houses, which is vital for maintaining optimal living conditions and ensuring the well-being of pigs. Temperature and humidity, along with concentrations of ammonia and hydrogen sulfide, are critical environmental factors that significantly influence the growth and health of pigs within porcine habitats. The ability to accurately predict these environmental variables in pig houses is pivotal, as it provides crucial decision-making support for the precise and targeted regulation of the internal environmental conditions. This approach ensures an optimal living environment, essential for the well-being and healthy development of the pigs. The existing methodologies for forecasting environmental factors in pig houses are currently hampered by issues of low predictive accuracy and significant fluctuations in environmental conditions. To address these challenges in this study, a hybrid model incorporating the improved dung beetle algorithm (DBO), temporal convolutional networks (TCNs), and gated recurrent units (GRUs) is proposed for the prediction and optimization of environmental factors in pig barns. The model enhances the global search capability of DBO by introducing the Osprey Eagle optimization algorithm (OOA). The hybrid model uses the optimization capability of DBO to initially fit the time-series data of environmental factors, and subsequently combines the long-term dependence capture capability of TCNs and the non-linear sequence processing capability of GRUs to accurately predict the residuals of the DBO fit. In the prediction of ammonia concentration, the OTDBO–TCN–GRU model shows excellent performance with mean absolute error (MAE), mean square error (MSE), and coefficient of determination (R2) of 0.0474, 0.0039, and 0.9871, respectively. Compared with the DBO–TCN–GRU model, OTDBO–TCN–GRU achieves significant reductions of 37.2% and 66.7% in MAE and MSE, respectively, while the R2 value is improved by 2.5%. Compared with the OOA model, the OTDBO–TCN–GRU achieved 48.7% and 74.2% reductions in the MAE and MSE metrics, respectively, while the R2 value improved by 3.6%. In addition, the improved OTDBO–TCN–GRU model has a prediction error of less than 0.3 mg/m3 for environmental gases compared with other algorithms, and has less influence on sudden environmental changes, which shows the robustness and adaptability of the model for environmental prediction. Therefore, the OTDBO–TCN–GRU model, as proposed in this study, optimizes the predictive performance of environmental factor time series and offers substantial decision support for environmental control in pig houses. [ABSTRACT FROM AUTHOR]

The power industry has been dedicated to applying virtual reality (VR) technology to build training systems in virtual environments, enabling personnel to complete skill training in real simulated environments while ensuring their safety. Conventional action scoring systems struggle to provide accurate scores for fine movements. Accurate scoring of fine movements can help workers identify their shortcomings during power operations, thus improving learning efficiency. This is of great significance for training on virtual environmentbased power operation. This paper proposes a power operationorientated VR action evaluation method based on the Gaussian distance-improved dynamic time warping (DTW) algorithm and the temporal convolutional network (TCN) model. First, the adaptive adapter is used to extract one-dimensional features from the three-dimensional data of the data gloves. Then, based on the TCN model, action data with significant discrepancies are filtered out. Finally, the obtained data are input into the Gaussian distance-improved DTW algorithm, where the path size is calculated. Corresponding scoring criteria are established on the basis of the path size to evaluate the actions. The results demonstrate that the VR action evaluation method based on the Gaussian distance-improved DTW algorithm and the TCN model significantly improves the accuracy of evaluating fine movements compared to traditional evaluation algorithms. [ABSTRACT FROM AUTHOR]

Non-intrusive load monitoring (NILM) is a technique which extracts individual appliance consumption and operation state change information from the aggregate power consumption made by a single residential or commercial unit. NILM plays a pivotal role in modernizing building energy management by disaggregating total energy consumption into individual appliance-level insights. This enables informed decision-making, energy optimization, and cost reduction. However, NILM encounters substantial challenges like signal noise, data availability, and data privacy concerns, necessitating advanced algorithms and robust methodologies to ensure accurate and secure energy disaggregation in real-world scenarios. Deep learning techniques have recently shown some promising results in NILM research, but training these neural networks requires significant labeled data. Obtaining initial sets of labeled data for the research by installing smart meters at the end of consumers' appliances is laborious and expensive and exposes users to severe privacy risks. It is also important to mention that most NILM research uses empirical observations instead of proper mathematical approaches to obtain the threshold value for determining appliance operation states (On/Off) from their respective energy consumption value. This paper proposes a novel semi-supervised multilabel deep learning technique based on temporal convolutional network (TCN) and long short-term memory (LSTM) for classifying appliance operation states from labeled and unlabeled data. The two thresholding techniques, namely Middle-Point Thresholding and Variance-Sensitive Thresholding, which are needed to derive the threshold values for determining appliance operation states, are also compared thoroughly. The superiority of the proposed model, along with finding the appliance states through the Middle-Point Thresholding method, is demonstrated through 15% improved overall improved F1micro score and almost 26% improved Hamming loss, F1 and Specificity score for the performance of individual appliance when compared to the benchmarking techniques that also used semi-supervised learning approach. [ABSTRACT FROM AUTHOR]

Copyright of China Rural Water & Hydropower is the property of China Rural Water & Hydropower Editorial Office and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Designing and developing photocatalysts with excellent performance in order to achieve efficient hydrogen production is an important strategy for addressing future energy and environmental challenges. Traditional single-phase photocatalytic materials either have a large bandgap and low visible light response or experience rapid recombination of the photogenerated carriers with low quantum efficiency, seriously hindering their photocatalytic applications. To solve these issues, an important solution is to construct wellmatched heterojunctions with highly efficient charge separation capabilities. To this end, an in situ sulfurization reaction was adopted after the deposition of Bi3+ supramolecular complex on a layered supramolecular precursor of tubular carbon nitride (TCN). X-ray diffraction (XRD) patterns confirmed that the asprepared sample has a good crystalline structure without any other impurities, while high-resolution transmission electron microscopy (HR-TEM) revealed that the heterojunction possesses a 2D structure with a layer of nano-array on its surface. Combined Fourier-transform infrared (FT-IR) spectra and energydispersive X-ray spectroscopy (EDX) revealed the interfacial interactions. Owing to the formation of the Z-scheme heterojunction, the visible light adsorption and the separation efficiency of the photo-generated carriers are both obviously enhanced, leaving the high energy electrons and high oxidative holes to participate in the photocatalytic reactions. As a result, the photocatalytic hydrogen evolution rate of Bi2S3-TCN achieves 65.2 µmol g-1]⋅h-1. This proposed green and environmentally benign route can also be applied to construct other sulfides with 2D TCN, providing some important information for the design and optimization of novel carbon-nitride-based semiconductors. [ABSTRACT FROM AUTHOR]

A large proportion of photovoltaic (PV) power generation is connected to the power grid, and its volatility and stochasticity have significant impacts on the power system. Accurate PV power forecasting is of great significance in optimizing the safe operation of the power grid and power market transactions. In this paper, a novel dual-channel PV power forecasting method based on a temporal convolutional network (TCN) is proposed. The method deeply integrates the PV station feature data with the model computing mechanism through the dual-channel model architecture; utilizes the combination of multihead attention (MHA) and TCN to extract the multidimensional spatio-temporal features between other meteorological variables and the PV power; and utilizes a single TCN to fully extract the temporal constraints of the power sequence elements. The weighted fusion of the dual-channel feature data ultimately yields the ideal forecasting results. The experimental data in this study are from a 26.52 kW PV power plant in central Australia. The experiments were carried out over seven different input window widths, and the two models that currently show superior performance within the field of PV power forecasting: the convolutional neural network (CNN), and the convolutional neural network combined with a long and short-term memory network (CNN_LSTM), are used as the baseline models. The experimental results show that the proposed model and the baseline models both obtained the best forecasting performance over a 1-day input window width, while the proposed model exhibited superior forecasting performance compared to the baseline model. It also shows that designing model architectures that deeply integrate the data input method with the model mechanism has research potential in the field of PV power forecasting. [ABSTRACT FROM AUTHOR]

Gross Domestic Product (GDP) is a good way to show how the economy is growing and how resources are being used in different areas. It's not enough to just look at a single factor when trying to figure out a region's GDP; you also need to look at factors like schooling, business, jobs, population, and more. It is important for both economics and government to be able to predict regional GDP because it tells us a lot about the economic health and growth of areas. We suggest using a Deep Reinforcement Learning-based multi-predictor ensemble decision system in this project to make area GDP forecasts more accurate and reliable. To train the GDP Prediction, we can use the Gated Recurrent Unit (GRU), the Temporal Convolutional Network (TCN), the Voting Classifier, the XGBooster, and the Deep Belief Network (DBN). These models can predict the data. In a multipredictior ensemble decision structure, a vote algorithm comes to a decision by putting together results from different predictors. The Deep Q-Network (DQN) method quickly checks how well these neural networks change to different GDP datasets so that an ensemble model can be made that gives correct results. [ABSTRACT FROM AUTHOR]

Wind speed exacerbates challenges associated with rock stability, introducing factors such as heightened erosion and the possibility of particle loosening. This increased sensitivity to erosion can result in material displacement, thereby compromising the overall stability of rock layers within the open-pit mining site. Therefore, accurate wind speed predictions are crucial for understanding the impact on rock stability, ensuring the safety and efficiency of open-pit mining operations. While most existing studies on wind speed prediction primarily focus on making overall predictions from the entire wind speed sequence, with limited consideration for the stationarity characteristics of the sequence, This paper introduces a novel approach for effective monitoring and early warning of geotechnical hazards. Our proposed method involves dividing wind speed data into stationary and non-stationary segments using the sliding window average method within the threshold method, validated by the Augmented Dickey-Fuller test. Subsequently, we use temporal convolutional networks (TCN) with dilated causal convolution and long short-term memory to predict the stationary segment of wind speed, effectively improving prediction accuracy for this segment. For the non-stationary segment, we implement complete ensemble empirical mode decomposition with adaptive noise (CEEMDAN) to reduce sequence complexity, followed by TCN with an attention mechanism (ATTENTION) to forecast wind speed one step ahead. Finally, we overlay the predictions of these two segments to obtain the final prediction. Our proposed model, tested with data from an open-pit mining area in western China, achieved promising results with an average absolute error of 0.14 knots, mean squared error of 0.05 knots2, and root mean squared error of 0.20 knots. These findings signify a significant advancement in the accuracy of short-term wind speed prediction. This advancement not oniy enables the rapid assessment and proactive response to imminent risks but also contributes to geotechnical hazard monitoring in open-pit mining operations. [ABSTRACT FROM AUTHOR]

Spark-Assisted Chemical Engraving (SACE) requires precise control over key factors to overcome gas film instability and achieve reproducible optimal resolution and machining speed. This paper presents a substantial advancement in the SACE micromanufacturing technique by introducing a composite algorithm. This algorithm leverages deep learning and time series classification for sequence-to-sequence intelligent classification. Two distinct architectures, the Temporal Convolutional Network (TCN) and Long Short-Term Memory (LSTM), were subjected to evaluation. They were trained and optimized using Bayesian optimization, resulting in impressive accuracies of 97.18% for TCN and 96.44% for LSTM. The algorithm relies on TCN due to its superior performance for classifying current data points and subsequently computing derived parameters like gas film formation time, lifetime, mean discharge current and energy. Its versatility is demonstrated across various experimental conditions, showcasing its potential for rapid and accurate systematic studies. By highlighting the algorithm's applicability in real-time process control for SACE, this study establishes a foundation for future advancements in the field of glass micromanufacturing. [ABSTRACT FROM AUTHOR]

Copyright of Computer Measurement & Control is the property of Magazine Agency of Computer Measurement & Control and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Abstract Physical systems whose dynamics are governed by partial differential equations (PDEs) find numerous applications in science and engineering. The process of obtaining the solution from such PDEs may be computationally expensive for large-scale and parameterized problems. In this work, deep learning techniques developed especially for time-series forecasts, such as LSTM and TCN, or for spatial-feature extraction such as CNN, are employed to model the system dynamics for advection-dominated problems. This paper proposes a Convolutional Autoencoder(CAE) model for compression and a CNN future-step predictor for forecasting. These models take as input a sequence of high-fidelity vector solutions for consecutive time steps obtained from the PDEs and forecast the solutions for the subsequent time steps using auto-regression; thereby reducing the computation time and power needed to obtain such high-fidelity solutions. Non-intrusive reduced-order modeling techniques such as deep auto-encoder networks are utilized to compress the high-fidelity snapshots before feeding them as input to the forecasting models in order to reduce the complexity and the required computations in the online and offline stages. The models are tested on numerical benchmarks (1D Burgers’ equation and Stoker’s dam-break problem) to assess the long-term prediction accuracy, even outside the training domain (i.e. extrapolation). The most accurate model is then used to model a hypothetical dam break in a river with complex 2D bathymetry. The proposed CNN future-step predictor revealed much more accurate forecasting than LSTM and TCN in the considered spatiotemporal problems.

Accurate and reliable PV power probabilistic-forecasting results can help grid operators and market participants better understand and cope with PV energy volatility and uncertainty and improve the efficiency of energy dispatch and operation, which plays an important role in application scenarios such as power market trading, risk management, and grid scheduling. In this paper, an innovative deep learning quantile regression ultra-short-term PV power-forecasting method is proposed. This method employs a two-branch deep learning architecture to forecast the conditional quantile of PV power; one branch is a QR-based stacked conventional convolutional neural network (QR_CNN), and the other is a QR-based temporal convolutional network (QR_TCN). The stacked CNN is used to focus on learning short-term local dependencies in PV power sequences, and the TCN is used to learn long-term temporal constraints between multi-feature data. These two branches extract different features from input data with different prior knowledge. By jointly training the two branches, the model is able to learn the probability distribution of PV power and obtain discrete conditional quantile forecasts of PV power in the ultra-short term. Then, based on these conditional quantile forecasts, a kernel density estimation method is used to estimate the PV power probability density function. The proposed method innovatively employs two ways of a priori knowledge injection: constructing a differential sequence of historical power as an input feature to provide more information about the ultrashort-term dynamics of the PV power and, at the same time, dividing it, together with all the other features, into two sets of inputs that contain different a priori features according to the demand of the forecasting task; and the dual-branching model architecture is designed to deeply match the data of the two sets of input features to the corresponding branching model computational mechanisms. The two a priori knowledge injection methods provide more effective features for the model and improve the forecasting performance and understandability of the model. The performance of the proposed model in point forecasting, interval forecasting, and probabilistic forecasting is comprehensively evaluated through the case of a real PV plant. The experimental results show that the proposed model performs well on the task of ultra-short-term PV power probabilistic forecasting and outperforms other state-of-the-art deep learning models in the field combined with QR. The proposed method in this paper can provide technical support for application scenarios such as energy scheduling, market trading, and risk management on the ultra-short-term time scale of the power system.

Temperature prediction is an important and significant step for monitoring global warming and the environment to save and protect human lives. The climatology parameters such as temperature, pressure, and wind speed are time-series data and are well predicted with data driven models. However, data-driven models have certain constraints, due to which these models are unable to predict the missing values and erroneous data caused by factors like sensor failure and natural disasters. In order to solve this issue, an efficient hybrid model, i.e., attention-based bidirectional long short term memory temporal convolution network (ABTCN) architecture is proposed. ABTCN uses k-nearest neighbor (KNN) imputation method for handling the missing data. A bidirectional long short term memory (Bi-LSTM) network with self-attention mechanism and temporal convolutional network (TCN) model that aids in the extraction of features from complex data and prediction of long data sequence. The performance of the proposed model is evaluated in comparison to various state-of-the-art deep learning models using error metrics such as MAE, MSE, RMSE, and R2 score. It is observed that our proposed model is superior over other models with high accuracy. [ABSTRACT FROM AUTHOR]

The measurement and analysis of vital signs are a subject of significant research interest, particularly for monitoring the driver's physiological state, which is of crucial importance for road safety. Various approaches have been proposed using contact techniques to measure vital signs. However, all of these methods are invasive and cumbersome for the driver. This paper proposes using a non-contact sensor based on continuous wave (CW) radar at 24 GHz to measure vital signs. We associate these measurements with distinct temporal neural networks to analyze the signals to detect and extract heart and respiration rates as well as classify the physiological state of the driver. This approach offers robust performance in estimating the exact values of heart and respiration rates and in classifying the driver's physiological state. It is non-invasive and requires no physical contact with the driver, making it particularly practical and safe. The results presented in this paper, derived from the use of a 1D Convolutional Neural Network (1D-CNN), a Temporal Convolutional Network (TCN), a Recurrent Neural Network particularly the Bidirectional Long Short-Term Memory (Bi-LSTM), and a Convolutional Recurrent Neural Network (CRNN). Among these, the CRNN emerged as the most effective Deep Learning approach for vital signal analysis. [ABSTRACT FROM AUTHOR]

Communication on the Controller Area Network (CAN) in vehicles is notably lacking in security measures, rendering it susceptible to remote attacks. These cyberattacks can potentially compromise safety-critical vehicle subsystems, and therefore endanger passengers and others around them. Identifying these intrusions could be done by monitoring the CAN traffic and detecting abnormalities in sensor measurements. To achieve this, we propose integrating time-series forecasting and signal correlation analysis to improve the detection accuracy of an onboard intrusion detection system (IDS). We predict sets of correlated signals collectively and report anomaly if their combined prediction error surpasses a predefined threshold. We show that this integrated approach enables the identification of a broader spectrum of attacks and significantly outperforms existing state-of-the-art solutions. [ABSTRACT FROM AUTHOR]

Fine particulate matter ( P M 2.5 ) poses a significant threat to human life and health, and therefore, accurately predicting P M 2.5 concentration is critical for controlling air pollution. Two improved types of recurrent neural networks (RNNs), the long short-term memory (LSTM) and gated recurrent unit (GRU), have been widely used in time series data prediction due to their ability to capture temporal features. However, both degrade into random guessing as the time length increases. In order to enhance the accuracy of P M 2.5 concentration prediction and address the issue of random guessing in RNNs neural networks, this study introduces a TCN-biGRU neural network model. This model is a hybrid prediction approach based on combining temporal convolutional networks (TCN) and bidirectional gated recurrent units (bi-GRU). TCN extracts higher-level feature information from longer time series data of P M 2.5 concentrations, while bi-GRU captures features from past and future data to achieve more accurate predictive outcomes. This case study utilizes data from monitoring stations in Beijing in 2021 for conducting P M 2.5 prediction experiments. The TCN-biGRU model achieves an average absolute error, root mean square error, and R 2 of 4.20, 7.71, and 0.961 in its predictive outcomes. When compared to the predictive outcomes of individual LSTM, GRU, and bi-GRU models, it is evident that the TCN-biGRU model exhibits smaller errors and superior predictive performance. [ABSTRACT FROM AUTHOR]

With the popularity of online learning and due to the significant influence of emotion on the learning effect, more and more researches focus on emotion recognition in online learning. Most of the current research uses the comments of the learning platform or the learner's expression for emotion recognition. The research data on other modalities are scarce. Most of the studies also ignore the impact of instructional videos on learners and the guidance of knowledge on data. Because of the need for other modal research data, we construct a synchronous multimodal data set for analyzing learners' emotional states in online learning scenarios. The data set recorded the eye movement data and photoplethysmography (PPG) signals of 68 subjects and the instructional video they watched. For the problem of ignoring the instructional videos on learners and ignoring the knowledge, a multimodal emotion recognition method in video learning based on knowledge enhancement is proposed. This method uses the knowledge-based features extracted from instructional videos, such as brightness, hue, saturation, the videos' clickthrough rate, and emotion generation time, to guide the emotion recognition process of physiological signals. This method uses Convolutional Neural Networks (CNN) and Long Short-TermMemory (LSTM) networks to extract deeper emotional representation and spatiotemporal information from shallow features. The model uses multi-head attention (MHA) mechanism to obtain critical information in the extracted deep features. Then, Temporal Convolutional Network (TCN) is used to learn the information in the deep features and knowledge-based features. Knowledge-based features are used to supplement and enhance the deep features of physiological signals. Finally, the fully connected layer is used for emotion recognition, and the recognition accuracy reaches 97.51%. Compared with two recent researches, the accuracy improved by 8.57% and 2.11%, respectively. On the four public data sets, our proposed method also achieves better results compared with the two recent researches. The experiment results show that the proposed multimodal emotion recognition method based on knowledge enhancement has good performance and robustness. [ABSTRACT FROM AUTHOR]

Lithium batteries have recently attracted significant attention as highly promising energy storage devices within the secondary battery industry. However, it is important to note that they may pose safety risks, including the potential for explosions during use. Therefore, achieving stable and safe utilization of these batteries necessitates accurate state-of-charge (SOC) estimation. In this study, we propose a hybrid model combining temporal convolutional network (TCN) and eXtreme gradient boosting (XGBoost) to investigate the nonlinear and evolving characteristics of batteries. The primary goal is to enhance SOC estimation performance by leveraging TCN's long-effective memory capabilities and XGBoost's robust generalization abilities. We conducted experiments using datasets from NASA, Oxford, and a vehicle simulator to validate the model's performance. Additionally, we compared the performance of our model with that of a multilayer neural network, long short-term memory, gated recurrent unit, XGBoost, and TCN. The experimental results confirm that our proposed TCN–XGBoost hybrid model outperforms the other models in SOC estimation across all datasets. [ABSTRACT FROM AUTHOR]

The vanadium content of molten iron is an important economic indicator for a vanadium–titanium magnetite smelting blast furnace, and it is of great importance in blast furnace production to be able to accurately predict it and optimize the operation of vanadium extraction. Based on the historical data of a commercial blast furnace, the clean data were obtained by processing the missing data and outlier data for data mining analysis and model development. A combined wavelet-TCN model was used to predict the vanadium content of molten iron. The average Hurst index after wavelet transform was calculated to reduce the complexity of the wavelet transform layer selection and the model computation time. The results show that compared to single models, such as LSTM, LSTM with attention, and TCN, the combined model based on wavelet-TCN (a = 5) had an improvement of about 11~17% in R2, and the prediction accuracy was high and stable, which met the practical requirements of blast furnace production. The factors affecting the vanadium content of molten iron were analyzed, and the measures to increase the vanadium content were summarized. A blast furnace should avoid increasing the titanium dioxide load, increase the vanadium load appropriately, and keep the relevant operating parameters within the appropriate range in order to achieve the optimization of vanadium extraction from molten iron. [ABSTRACT FROM AUTHOR]

A multi-strategy improved sand cat swarm algorithm with PTCN-BIGRU is proposed to solve the problem of predicting the risk level of gas-containing coal fracture. Combined with kernel entropy component analysis to downscale the gas-containing coal fracture risk level predictors, TCN is used for feature extraction by parallel convolution operation, and BiGRU is used to further obtain the contextual links of the features. A parameterized exponential linear unit based on the standard TCN is used to improve the linear unit and to enhance the generalization capability of the model. Combined with the sand cat swarm optimization algorithm to determine the optimal BIGRU network parameters, Singer chaos mapping, chaos decreasing factor, and adaptive t-distribution are used to improve the SCSO for optimal risk level prediction accuracy. The results show that the prediction accuracy of the ISCSO-PTCN-BiGRU model is 93.33%, which is better than other models, and it is proved that this paper can effectively improve the prediction accuracy of gas-containing coal fracture risk level. This research adds a theoretical support for the prevention of gas protrusion accidents and a guarantee for the safety of underground production in coal mines. [ABSTRACT FROM AUTHOR]

Student engagement is an important factor in meeting the goals of virtual learning programs. Automatic measurement of student engagement provides helpful information for instructors to meet learning program objectives and individualize program delivery. Many existing approaches solve video-based engagement measurement using the traditional frameworks of binary classification (classifying video snippets into engaged or disengaged classes), multi-class classification (classifying video snippets into multiple classes corresponding to different levels of engagement), or regression (estimating a continuous value corresponding to the level of engagement). However, we observe that while the engagement behavior is mostly well defined (e.g., focused, not distracted), disengagement can be expressed in various ways. In addition, in some cases, the data for disengaged classes may not be sufficient to train generalizable binary or multi-class classifiers. To handle this situation, in this paper, for the first time, we formulate detecting disengagement in virtual learning as an anomaly detection problem. We design various autoencoders, including temporal convolutional network autoencoder, long short-term memory autoencoder, and feedforward autoencoder using different behavioral and affect features for video-based student disengagement detection. The result of our experiments on two publicly available student engagement datasets, DAiSEE and EmotiW, shows the superiority of the proposed approach for disengagement detection as an anomaly compared to binary classifiers for classifying videos into engaged versus disengaged classes (with an average improvement of 9% on the area under the curve of the receiver operating characteristic curve and 22% on the area under the curve of the precision–recall curve). [ABSTRACT FROM AUTHOR]

Inferring the remaining useful life (RUL) of an aero-engine based on complex data from aircraft sensors is one of the important issues to ensure flight safety. To this end, this paper is intended to propose a RUL prediction model based on the feature extraction method and the improved temporal convolution network (TCN). First, the XGBoost (eXtreme Gradient Boosting) model is used to assess the importance of the data and filter the features base on the resulting correlation. Then, the RUL prediction model is constructed by paralleling TCN networks with different expansion rates, which expands the receptive field and further improves the information obtained by the network from the data. Moreover, the network is further optimized with dynamic hyperparameter search methods. Finally, through comparative experiments, the proposed prediction model is evaluated based on the turbofan aero-engine operation failure prediction benchmark dataset (CMAPSS). The experimental results show that by deleting some data with low correlation, the proposed model can achieve better prediction accuracy, which is superior to other mainstream models in the references.