Monaural Speech Enhancement Using a Multi-Branch Temporal Convolutional Network

Veg-W2TCN: A parallel hybrid forecasting framework for non-stationary time series using wavelet and temporal convolution network model

Abstract. The present work proposes a prediction model of significant wave height (SWH) and average wave period (APD) based on variational mode decomposition (VMD), temporal convolutional networks (TCNs), and long short-term memory (LSTM) networks. The wave sequence features were obtained using VMD technology based on the wave data from the National Data Buoy Center. Then the SWH and APD prediction models were established using TCNs, LSTM, and Bayesian hyperparameter optimization. The VMD–TCN–LSTM model was compared with the VMD–LSTM (without TCN cells) and LSTM (without VMD and TCN cells) models. The VMD–TCN–LSTM model has significant superiority and shows robustness and generality in different buoy prediction experiments. In the 3 h wave forecasts, VMD primarily improved the model performance, while the TCN had less of an influence. In the 12, 24, and 48 h wave forecasts, both VMD and TCNs improved the model performance. The contribution of the TCN to the improvement of the prediction result determination coefficient gradually increased as the forecasting length increased. In the 48 h SWH forecasts, the VMD and TCN improved the determination coefficient by 132.5 % and 36.8 %, respectively. In the 48 h APD forecasts, the VMD and TCN improved the determination coefficient by 119.7 % and 40.9 %, respectively.

Accurate and reliable estimation of reference crop evapotranspiration (ET0) in the North Henan Plain is crucial for agricultural water resource management, production, and food supply in China. This study aims to evaluate the performance of deep learning (DL) methods in ET0 estimation and assess the applicability of the developed DL model beyond the training domain. This study utilized historical meteorological data from Zhengzhou City, northern Henan, spanning 2010–2024. Meteorological variables were selected through correlation analysis and maximum information coefficient (MIC). A novel DL model—the TCN-Attention model (TA)—was constructed by incorporating a self-attention mechanism into the temporal convolutional network (TCN) model. This model was compared with two classical DL models—Long Short-Term Memory (LSTM) and TCN. Results indicate: (1) Sunshine duration (n), relative humidity (RH), and maximum temperature (Tmax) are the three most significant features influencing summer maize evapotranspiration; (2) prediction accuracy under the same input scenarios: TA model > TCN model > LSTM model; (3) in scenarios where only temperature data is input, the TA model has the highest prediction accuracy, surpassing the H-S empirical method; and (4) for limited meteorological data, the combination of temperature and humidity was found to be most effective, showing good adaptability and accuracy at different time steps (hourly: R2 = 0.982; daily: R2 = 0.975; weekly: R2 = 0.928). This study highlights the potential of the TA model for estimating reference crop evapotranspiration in the northern Henan Plain, which may provide theoretical guidance for crop irrigation management under future climate change.

A deep learning framework for constitutive modeling based on temporal convolutional network

Masked multi-head self-attention for causal speech enhancement

When considering agricultural commodity transaction data, long sampling intervals or data sparsity may lead to small samples. Furthermore, training on small samples can lead to overfitting and makes it hard to capture the fine-grained fluctuations in the data. In this study, a multi-scale forecasting approach combined with a Generative Adversarial Network (GAN) and Temporal Convolutional Network (TCN) is proposed to address the problems related to small sample prediction. First, a Time-series Generative Adversarial Network (TimeGAN) is used to expand the multi-dimensional data and t-SNE is utilized to evaluate the similarity between the original and synthetic data. Second, a greedy algorithm is exploited to calculate the information gain, in order to obtain important features, based on XGBoost. Meanwhile, TCN residual blocks and dilated convolutions are used to tackle the issue of gradient disappearance. Finally, an attention mechanism is added to the TCN, which is beneficial in terms of improving the forecasting accuracy. Experiments are conducted on three products, garlic, ginger and chili. Taking garlic as an example, the RMSE of the proposed method was reduced by 1.7% and 1% when compared to the SVR and RF models, respectively. Its R2 accuracy was also improved (by 4.3% and 3.4%, respectively). Furthermore, TCN-attention and TCN were found to require less time compared to GRU and LSTM. The accuracy of the proposed method increased by about 5% when compared to that without TimeGAN in the ablation study. Moreover, compared with TCN, the Gated Recurrent Unit (GRU), and the Long Short-term Memory (LSTM) model in the multi-scale price forecasting task, the proposed method can better utilize small samples and high-dimensional data, leading to improved performance. Additionally, the proposed model is compared to the Transformer and TimesNet models in terms of its accuracy, deployment cost, and other metrics.

Multivariate time-series forecasting is one of the crucial and persistent challenges in time-series forecasting tasks. As a kind of data with multivariate correlation and volatility, multivariate time series impose highly nonlinear time characteristics on the forecasting model. In this paper, a new multivariate time-series forecasting model, multivariate temporal convolutional attention network (MTCAN), based on a self-attentive mechanism is proposed. MTCAN is based on the Convolution Neural Network (CNN) model, using 1D dilated convolution as the basic unit to construct asymmetric blocks, and then, the feature extraction is performed by the self-attention mechanism to finally obtain the prediction results. The input and output lengths of this network can be determined flexibly. The validation of the method is carried out with three different multivariate time-series datasets. The reliability and accuracy of the prediction results are compared with Long Short-Term Memory (LSTM), Gated Recurrent Unit (GRU), Convolutional Long Short-Term Memory (ConvLSTM), and Temporal Convolutional Network (TCN). The prediction results show that the model proposed in this paper has significantly improved prediction accuracy and generalization.

Hybrid solar radiation forecasting model with temporal convolutional network using data decomposition and improved artificial ecosystem-based optimization algorithm

In the realm of power systems, short-term electric load forecasting is pivotal for ensuring supply–demand balance, optimizing generation planning, reducing operational costs, and maintaining grid stability. Short-term load curves are characteristically coarse, revealing high-frequency data upon decomposition that exhibit pronounced non-linearity and significant noise, complicating efforts to enhance forecasting precision. To address these challenges, this study introduces an innovative model. This model employs complete ensemble empirical mode decomposition with adaptive noise (CEEMDAN) to bifurcate the original load data into low- and high-frequency components. For the smoother low-frequency data, a temporal convolutional network (TCN) is utilized, whereas the high-frequency components, which encapsulate detailed load history information yet suffer from a lower fitting accuracy, are processed using an enhanced soft thresholding TCN (SF-TCN) optimized with the slime mould algorithm (SMA). Experimental tests of this methodology on load forecasts for the forthcoming 24 h across all seasons have demonstrated its superior forecasting accuracy compared to that of non-decomposed models, such as support vector regression (SVR), recurrent neural network (RNN), gated recurrent unit (GRU), long short-term memory (LSTM), convolutional neural network-LSTM (CNN-LSTM), TCN, Informer, and decomposed models, including CEEMDAN-TCN and CEEMDAN-TCN-SMA.

Accurate short-term GHI forecasting using a novel temporal convolutional network model

Research on smartphone-based human activity recognition (HAR) is prevalent in the field of healthcare, especially for elderly activity monitoring. Researchers usually propose to use of accelerometers, gyroscopes or magnetometers that are equipped in smartphones as an individual sensing modality for human activity recognition. However, any of these alone is limited in capturing comprehensive movement information for accurate human activity analysis. Thus, we propose a smartphone-based HAR approach by leveraging the inertial signals captured by these three sensors to classify human activities. These heterogeneous sensors deliver information on various aspects of nature, motion and orientation, offering a richer set of features for more accurate representations of the activities. Hence, a deep learning approach that amalgamates long short-term memory (LSTM) in temporal convolutional network (TCN) is proposed. We use independent temporal convolutional networks, coined as temporal convolutional streams, to independently analyse the temporal data of each sensing modality. We name this architecture multi-stream TC-LSTM. The performance of multi-stream TC-LSTM is assessed on the self-collected elderly activity database. Empirical results exhibit that multi-stream TC-LSTM outperforms the existing machine learning and deep learning models, with an F1 score of 98.3 %

A unified modeling framework for lane change intention recognition and vehicle status prediction

Forecasting photovoltaic (PV) power generation, as in many other time series scenarios, is a challenging task. Most current solutions for time series forecasting are grounded on Machine Learning (ML) algorithms, which usually outperform statistical-based methods. However, solutions based on ML and, more recently, Deep Learning (DL) have been found vulnerable to adversarial attacks throughout their execution. With this in mind, in this work we explore four time series analysis techniques, namely Naive, a baseline technique for time series, Auto-regressive Integrated Moving Average (ARIMA), from the statistical field, and Long Short-term Memory (LSTM) and Temporal Convolutional Network (TCN), from the DL family. These techniques are used to forecast the power generation of a PV power plant 15 minutes and 24 hours ahead, having as input only power generation historical data. Two main aspects were analyzed: i) how training size influenced the performance of the forecasting models and ii) how univariate time series data could be modified by an adversarial attack to decrease models’ performance through cross-technique transferability. For i), the mentioned methods were used and evaluated with monthly updates. For ii), Fast Gradient Sign Method (FGSM), along with a logistic regression substitute model and past data, were used to perform attacks against DL models at test time. LSTM and TCN decreased the error as the training sample size increased and outperformed Naive and ARIMA models. Adversarial samples were able to reduce the performance of LSTM and TCN, particularly for short-term forecasts.

The prediction of wind energy generation is important to enhance the performance and dependability of renewable energy systems due to the rising demand for wind-generated electricity and advancements in wind energy technology competitiveness. This study leverages advanced machine learning (ML) and some other statistical and deep learning based time series forecasting models to enhance the accuracy of wind energy predictions. This comprehensive analysis includes nine ML models—Linear Regression, Random Forests (RF), Gradient Boosting Machines (GBM), Support Vector Machines (SVM), K-Nearest Neighbors (KNN), AdaBoost, XGBoost, Support Vector Regression (SVR), and Neural Networks—as well as Four time-series forecasting models—ARIMA, Temporal Convolutional Networks (TCNs), Long Short-Term Memory (LSTM) networks and GRU. Each ML model underwent rigorous cross-validation to ensure optimal performance. The assessment criteria utilized here comprised the Mean Squared Error (MSE), Root Mean Squared Error (RMSE), Mean Absolute Error (MAE), and the R² Score. It was found that among the nine ML models, Random Forests, GBM and KNN consistently provided superior accuracy and robustness, making them the top choices for wind energy prediction whereas the performance of linear regression, SVM and SVR were very poor for the considered dataset. From the experiment, Random Forest, GBM, and KNN showed the best performance with low MSE values of 0.77, 1.95, and 1.51 respectively, while other models had MSEs above 7.5, with AdaBoost reaching 30. Their RMSEs (0.88, 1.40, 1.23) and MAEs (0.093, 0.73, 0.10) also indicate strong predictive accuracy compared to the rest.In this paper, time series forecasting, TCNs, LSTM and GRU networks showed strong capabilities in capturing temporal dependencies and trends within the wind energy data. Visualization techniques were employed to compare model performances comprehensively, providing clear insights into their predictive power. Therefore, this present study offers a robust framework for researchers and practitioners aiming to leverage machine learning and time series forecasting in the realm of renewable energy prediction.

This study aimed to improve apple slices’ color and drying kinetics by optimizing the hot-air drying process, utilizing machine and deep learning models. Different steam blanching times (30, 60, 90, and 120 s), drying temperatures (50, 55, 60, 65, and 70 °C), and humidity control methods (full humidity removal or temperature–humidity control) were examined. These factors significantly affected the quality of apple slices. 60 s blanching, 60 °C temperature, and full dehumidification represented the optimal drying conditions for apple slices’ dehydration, achieving better drying kinetics and the best color quality. However, the fastest drying process (40 min) was obtained at a 60 °C drying temperature combined with complete dehumidification after 90 s blanching. Furthermore, machine and deep learning models, including backpropagation (BP), convolutional neural network–long short-term memory (CNN-LSTM), temporal convolutional network (TCN), and long short-term memory (LSTM) networks, effectively predicted the moisture content and color variation in apple slices. Among these, LSTM networks demonstrated exceptional predictive performance with an R2 value exceeding 0.98, indicating superior accuracy. This study provides a scientific foundation for optimizing the drying process of apple slices and illustrates the potential application of deep learning in the agricultural processing and engineering fields.