Temporal convolutional network-based electromagnetic power spectrum density prediction 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.

To accurately predict the life of proton exchange membrane fuel cells (PEMFC), the IGJO-TCN-BiGRU-Attention algorithm is proposed, which integrates an intelligent optimization algorithm into a deep learning framework. Temporal convolutional network (TCN) decodes spatial correlations in voltage-temperature parameters. Bidirectional gated recurrent units (BiGRU) models performance degradation dynamics. Attention mechanism makes the model focus on the part that contributes the most to the predicted outcome. The improved golden jackal optimization (IGJO) algorithm proposed in this work is employed to optimize the learning rate, L2 regularization parameters, number of BiGRU neurons, and key-value pairs of the attention mechanism in the TCN-BiGRU-Attention hybrid model. Subsequently, the optimized hybrid model is utilized to predict the life of PEMFC. Superior convergence speed and optimization accuracy of the proposed IGJO algorithm are validated through comparative simulations with five intelligent algorithms. Further validation under static and dynamic operating conditions with varying training durations exhibits significantly enhanced prediction accuracy of the IGJO-TCN-BiGRU-Attention algorithm. Under static operating conditions at 577 hours, a 67.78% reduction in root mean square error (RMSE) is achieved compared to TCN. Similarly, under dynamic conditions at 510 hours, a 22.5% RMSE reduction is observed versus TCN.

BackgroundThe dynamic tracking of tumors with radiation beams in radiation therapy requires the prediction of real-time target locations prior to beam delivery, as treatment involving radiation beams and gating tracking results in time latency.ObjectiveIn this study, a deep learning model that was based on a temporal convolutional neural network was developed to predict internal target locations by using multiple external markers.MethodsRespiratory signals from 69 treatment fractions of 21 patients with cancer who were treated with the CyberKnife Synchrony device (Accuray Incorporated) were used to train and test the model. The reported model’s performance was evaluated by comparing the model to a long short-term memory model in terms of the root mean square errors (RMSEs) of real and predicted respiratory signals. The effect of the number of external markers was also investigated.ResultsThe average RMSEs of predicted (ahead time=400 ms) respiratory motion in the superior-inferior, anterior-posterior, and left-right directions and in 3D space were 0.49 mm, 0.28 mm, 0.25 mm, and 0.67 mm, respectively.ConclusionsThe experiment results demonstrated that the temporal convolutional neural network–based respiratory prediction model could predict respiratory signals with submillimeter accuracy.

Drought constitutes one of the most significant natural hazards worldwide, exacerbated by climate variability and change, with profound implications for ecosystems, agriculture, and livelihoods. In South Africa, particularly within the drought-prone uMkhanyakude District of KwaZulu-Natal, comprehending rainfall variability and enhancing drought prediction are imperative for sustainable water and food security planning. This study utilized daily rainfall records from six meteorological stations spanning the years 1980 to 2023 to calculate the Standardized Precipitation Index (SPI) at 6-, 9-, and 12-month time scales. Long-term drought trends were evaluated employing Innovative Trend Analysis (ITA) methods, which identified statistically significant decreasing trends at five stations and an increasing trend at Riverview. To augment drought forecasting, a novel hybrid model that integrates the Savitzky-Golay filter with a Temporal Convolutional Network and Long Short-Term Memory (SG-TCN-LSTM) was developed. Comparative assessments against ARIMA, LSTM, TCN, and other hybrid models demonstrated that the SG-TCN-LSTM consistently achieved the lowest Root Mean Square Error (RMSE) values (0.0349-0.1453) and the highest [Formula: see text] values (0.95-0.99) across all SPI scales, indicating superior predictive accuracy and stability. The integration of signal smoothing with deep learning methodologies enhanced the robustness of forecasts, providing critical insights for proactive drought risk management. This research underscores the potential of hybrid models as reliable early-warning instruments for meteorological drought and establishes a framework that can inform national adaptation strategies. Future research should aim to extend the model to incorporate additional climatic drivers, evaluate its transferability across diverse climatic regions, and investigate its application in operational drought early-warning systems.

Multisource biosignals are being increasingly used in human–machine interaction applications. In particular, a critical problem in exoskeleton-assisted rehabilitation for patients with hemiplegia is the generation of rhythmic and symmetrical locomotion. To support lower limb rehabilitation, we propose a model for predicting contralateral joint angles using multisource biosignals. First, a vibroarthrography (VAG) sensor is attached to the affected leg, and surface electromyography sensors are attached to the sound leg. The corresponding signals are used to estimate the hip, knee, and ankle joint angles of the affected leg. Second, an algorithm based on a temporal convolution network (TCN) is introduced to predict the contralateral lower-limb joint angles during human locomotion. The TCN is compared with a long short-term memory (LSTM) network and a convolutional neural network. Experiments were conducted by 10 healthy participants. The results of the proposed model were compared with measurements from encoders in three joints mounted on an exoskeleton to verify the applicability of the proposed model. In addition, by using the outputs of a motion capture system as the ground truth, the experimental results validated the model prediction performance. The prediction root mean square error (RMSE) of the TCN was 52%–70% lower than that of the LSTM network and CNN at different paces. The prediction RMSE using VAG was 20%–24% lower than that without using VAG. Note to Practitioners—We propose a model for predicting the contralateral lower-limb joint angles of an exoskeleton. The model uses measurements from multiple biosignal sensors to support exoskeleton-assisted hemiplegia rehabilitation. A sensing system based on acoustic signals and bioelectric signals is developed to improve the joint angle prediction in the affected leg of patients with hemiplegia. A temporal convolution network is implemented to estimate joint angles, and training considering the root mean square error and Pearson correlation coefficient as evaluation indicators is performed. Unlike similar methods that include no sensory feedback from the affected leg, the proposed method incorporates VAG signals from the affected leg and sEMG signals from the sound leg to handle different degrees of hemiplegia and support the generation of rhythmic and symmetrical locomotion for gait rehabilitation using an exoskeleton.

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.

The accuracy of short-term electricity price forecasting has profound influence on the business decisions made by the participants in the electricity market. Thus, we propose a short-term electricity price combination prediction model based on Complete Ensemble Empirical Mode Decomposition(CEEMD) as well as Temporal Convolutional Attention-based Network(TCAN). First, we defined the correlations among the features according to the Pearson Correlation Analysis and selected the desired features for our model. We then defined the single prediction period to be 24 hours based on the Unit Root test. In addition, we decomposed and denoised the data based on CEEMD and built the attention mechanism into the temporal convolution network. Using the expansion cascade network to extract the relationship between the inputs, combined with the self-attention mechanism to extract the internal information and learn the position dependence, effectively improve the model training speed and prediction accuracy. In order to verify the validity of the model, the actual electricity price data of DK1 in the Nordic electricity market from April to May 2020 were forecasted. The results show that the proposed method can effectively improve the accuracy and reliability of electricity price forecasting, and can provide reliable decision-making basis.

Battery health prognosis using improved temporal convolutional network modeling

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

Seeking green and low-carbon development has become a common consensus of the international community, and wind energy is one of the best new energy sources for the development of low-carbon economy. However, wind power has strong randomness and volatility, which will bring new challenges to the green and low-carbon operation of new power systems. Relying on the new power system, accurate power prediction of wind turbines can not only optimize the power supply structure of the power system, but also promote energy conservation, emission reduction and ‘low-carbon’ development of the power industry. Therefore, under the above background, this paper proposes a short-term wind power prediction model based on Singular Spectrum Analysis-Temporal Convolutional Network-Support Vector Machine (SSA-TCN-SVM). Aiming at the problem that the Temporal Convolutional Network (TCN) prediction model still has complex modes and the Dense Layer cannot set the weights reasonably, a cascade model based on Temporal Convolutional Network-Support Vector Machine (TCN-SVM) is adopted. The kernel function of Support Vector Machine (SVM) is used to map the vectors output by the convolution kernel channel of each time to the high-dimensional kernel space. The optimal weights are obtained by solving the dual problem, and the prediction ability of the single grouping sequence prediction model is improved. By minimizing the prediction error of the grouping sequence, the prediction accuracy of the whole prediction model is improved. Experiments show that SSA-TCN-SVM has better prediction performance.

The precise and real-time forecasting of traffic is very important for the planning, control, and orientation of traffic in cities. However, forecasting circulation flows remain a troublesome issue because of the highly non-linear and complex nature of circulation systems. This research proposes a new temporal convolutional multi-attentive network-based traffic flow prediction model (TCMAN). TCMAN model captures the spatiotemporal features of traffic flow through temporal convolutional network (TCN) and codec. The codec includes several spatial-temporal attention blocks to simulate the effects of space-time factors on circulation conditions. The input traffic flow features are coded by the encoder and the output sequence is anticipated by the decoder. Finally, many experiments are carried out on the traffic datasets. Compared with the baseline method, TCMAN model has better prediction performance.

Due to the inherent high volatility and complexity of financial markets, traditional time series forecasting models face numerous challenges in handling both short- and long-term predictions in the stock market. Most traditional neural network-based financial prediction models are limited to short-term forecasting and struggle to capture long-term trends and global dependencies in the market fully. To address this, we propose a novel network architecture called ResMMoT-Informer. This model combines the strengths of the residual multiscale temporal convolutional network (TCN) sparse expert network (ResMMoT) and the Informer, enabling it to effectively capture multiscale local features and global dependencies in the stock market. ResMMoT achieves stable training through a residual structure and a sparse multiscale TCN expert network, allowing it to flexibly model complex temporal features and learn trends across different time-step scales. Meanwhile, the Informer optimizes long-sequence forecasting performance through an improved self-attention mechanism. Additionally, we introduce the wavelet noise reduction (WNR) method, further enhancing the model's robustness and prediction accuracy. In the experimental section, ablation experiments first validate the effectiveness and necessity of the proposed strategies and network structure. Subsequent comparison experiments on the NASDAQ100 dataset demonstrate that ResMMoT-Informer excels in both long- and short-term time series forecasting tasks in the stock market, with significantly better prediction accuracy and generalization ability than existing models. Compared to other popular neural network-based financial forecasting models, ResMMoT-Informer leads in prediction accuracy, time robustness, and interpretability, showcasing its cutting-edge advantage in contemporary research.

Natural gas is often said to be the most environmentally friendly fossil fuel. Its usage has increased significantly in recent years. Meanwhile, accurate forecasting of natural gas spot prices has become critical to energy management, economic growth, and environmental protection. This work offers a novel model based on the temporal convolutional network (TCN) and dynamic learning rate for predicting natural gas spot prices over the following two weekdays. The residual block structure of TCN provides good prediction accuracy, and the dilated causal convolutions minimize the amount of computation. The dynamic learning rate setting was adopted to enhance the model’s prediction accuracy and robustness. Compared with three existing models, i.e., the one-dimensional convolutional neural network (1D-CNN), gate recurrent unit (GRU), and long short-term memory (LSTM), the proposed model can achieve better performance over other models with mean absolute percentage error (MAPE), mean absolute error (MAE), and root mean squared error (RMSE) scores of 4.965%, 0.216, and 0.687, respectively. These attractive advantages make the proposed model a promising candidate for long-term stability in natural gas spot price forecasting.

Purpose The study aims to develop a deep learning-based predictive framework for accurately forecasting transient residual stress profiles in the directed energy deposition (DED) process, thereby addressing a critical challenge in metal additive manufacturing. Design/methodology/approach Time-series data generated from experimentally validated finite element simulations were employed to train and evaluate four temporal deep learning models, including LSTM, GRU, Bi-LSTM and temporal convolutional networks (TCN). Findings Among the models tested, the TCN architecture achieved the best performance, with an R2 of 0.99 and a root mean square error (RMSE) of 4.7 MPa. This demonstrates its capability to capture complex temporal variations in residual stress with high fidelity. Research limitations/implications Applicable for single track deposition. Practical implications It can be utilized as an efficient for in situ monitoring of transient residual stress. Social implications Utilized for minimizing computational and experimental fabrication cost. Originality/value The results establish TCN as a robust and scalable approach for real-time residual stress prediction in DED. This advancement provides a pathway for improved process control and optimization, ultimately enhancing the reliability and quality of additively manufactured metal components.

The current prediction model for the system efficiency of pumping units primarily relies on a mechanistic approach. However, this approach incorporates numerous unnecessary factors, thereby, increasing the cost associated with predictions. With the improvement of the oil field database, the available information is increasing. Some scholars propose a prediction model based on a single neural network, however, such models face challenges in effectively capturing complex data, resulting in lower prediction accuracy and limited resistance to interference. To solve the above problems, the study proposes a novel stacking integrated learning prediction model, which incorporates fivefold cross‐validation. First, the magnitude of the correlation coefficient was quantified using the Pearson correlation coefficient. Second, the impact and predictive features were normalized. Final, convolutional neural network (CNN), recurrent neural network (RNN), Long Short‐Term Memory network (LSTM), gated recurrent unit (GRU), and transformer are used as the base models, and fully connected neural network (FNN) is used as the metamodel. Each base model was trained by fivefold cross‐validation, and the predicted values of each fold were stacked by rows. Next, the predicted values of each base model are stacked by columns as input variables to the metamodel and metamodel learning is performed, and the stacking integrated learning prediction model based on fivefold crossover validation is established. To validate the accuracy of the model, we selected 5,000 actual well parameters, including 26 impact features and one predictive feature, for comparative analysis. This analysis presents the maximum percentage reduction in mean square error (MSE), mean absolute error (MAE), and root‐mean‐square error (RMSE) of our proposed integrated learning model concerning a single neural network prediction model as 28.26%, 24.40%, and 15.66%, respectively. The maximum percentage improvement in R2 is 17.74%. It shows that our proposed integrated learning prediction model has high prediction accuracy.