Conventional Model Research Articles

SR-TTS: a rhyme-based end-to-end speech synthesis system.

Deep learning has significantly advanced text-to-speech (TTS) systems. These neural network-based systems have enhanced speech synthesis quality and are increasingly vital in applications like human-computer interaction. However, conventional TTS models still face challenges, as the synthesized speeches often lack naturalness and expressiveness. Additionally, the slow inference speed, reflecting low efficiency, contributes to the reduced voice quality. This paper introduces SynthRhythm-TTS (SR-TTS), an optimized Transformer-based structure designed to enhance synthesized speech. SR-TTS not only improves phonological quality and naturalness but also accelerates the speech generation process, thereby increasing inference efficiency. SR-TTS contains an encoder, a rhythm coordinator, and a decoder. In particular, a pre-duration predictor within the cadence coordinator and a self-attention-based feature predictor work together to enhance the naturalness and articulatory accuracy of speech. In addition, the introduction of causal convolution enhances the consistency of the time series. The cross-linguistic capability of SR-TTS is validated by training it on both English and Chinese corpora. Human evaluation shows that SR-TTS outperforms existing techniques in terms of speech quality and naturalness of expression. This technology is particularly suitable for applications that require high-quality natural speech, such as intelligent assistants, speech synthesized podcasts, and human-computer interaction.

Machine Learning-Based Interpretable Modeling for Subjective Emotional Dynamics Sensing Using Facial EMG.

Understanding the association between subjective emotional experiences and physiological signals is of practical and theoretical significance. Previous psychophysiological studies have shown a linear relationship between dynamic emotional valence experiences and facial electromyography (EMG) activities. However, whether and how subjective emotional valence dynamics relate to facial EMG changes nonlinearly remains unknown. To investigate this issue, we re-analyzed the data of two previous studies that measured dynamic valence ratings and facial EMG of the corrugator supercilii and zygomatic major muscles from 50 participants who viewed emotional film clips. We employed multilinear regression analyses and two nonlinear machine learning (ML) models: random forest and long short-term memory. In cross-validation, these ML models outperformed linear regression in terms of the mean squared error and correlation coefficient. Interpretation of the random forest model using the SHapley Additive exPlanation tool revealed nonlinear and interactive associations between several EMG features and subjective valence dynamics. These findings suggest that nonlinear ML models can better fit the relationship between subjective emotional valence dynamics and facial EMG than conventional linear models and highlight a nonlinear and complex relationship. The findings encourage emotion sensing using facial EMG and offer insight into the subjective-physiological association.

Rate dependent cohesive zone model for fatigue crack growth

Experimental observations indicate that the frequency of cyclic loads can influence fatigue crack growth. To address this phenomenon in simulation modelling, this study introduces a rate-dependent cohesive zone model designed for analysing fatigue crack growth under arbitrary cyclic loading conditions. The model operates in the time domain, where both critical traction and fracture toughness are modelled as functions of the separation rate, thereby considering the rate dependency of material damage and cracking. Unlike conventional models, this approach does not rely directly on loading frequency, suggesting its capability to address fatigue crack growth under non-periodic loading or repeated impacts effectively. An algorithm for modelling fatigue crack evolution and growth was developed based on this model, employing a time-incremental strategy in conjunction with a linear cumulative damage method. This algorithm was implemented in ABAQUS using a user element subroutine (UEL). The effectiveness of the proposed model and algorithm is validated through examples of fatigue crack growth in a plate under cyclic loading with varying frequencies. The results demonstrate agreement with experimental and computational findings reported in the literature. Additionally, the model was applied to scenarios involving frequency-variable cyclic loadings, yielding reasonable outcomes. Analysis of these examples confirms the model's utility in predicting fatigue crack growth under diverse alternating load conditions.

Application of Gamification Models with Virtual Reality for Learning Plant Cultivation Techniques

The aim of this research is to assess the effectiveness of implementing gamification models with virtual reality (VR) for plant cultivation techniques in constructing hydroponic circuits. The purpose of this study is to examine the significant differences in effectiveness between the control group and the intervention group when using gamification with virtual reality. The data used consists of test results from 50 respondents, divided into two groups: a control group of 25 students and an intervention group of 25 students. The trial was divided into two different groups. Group 1, the control group, received conventional treatment, which involved lecturers explaining the assembly of hydroponic circuits in class, followed by students practicing the assembly. Group 2, the intervention group, received self-learning treatment that involved the practice of assembling hydroponic circuits using a gamification model with VR. An independent t-test was conducted to determine whether there was a significant effect of applying plant cultivation techniques on assembling hydroponic circuits. The results of the t-test revealed a significant effect of applying the gamification model with VR compared to the conventional model. The application of the gamification model with VR to assemble hydroponic circuits has shown significant effectiveness, with an N-Gain value of 62.47%, whereas the conventional treatment has demonstrated lower effectiveness, with an N-Gain value of 28.24%.

A hybrid machine learning-based model for predicting flight delay through aviation big data

The prediction of flight delays is one of the important and challenging issues in the field of scheduling and planning flights by airports and airlines. Therefore, in recent years, we have witnessed various methods to solve this problem using machine learning techniques. In this article, a new method is proposed to address these issues. In the proposed method, a group of potential indicators related to flight delay is introduced, and a combination of ANOVA and the Forward Sequential Feature Selection (FSFS) algorithm is used to determine the most influential indicators on flight delays. To overcome the challenges related to large flight data volumes, a clustering strategy based on the DBSCAN algorithm is employed. In this approach, samples are clustered into similar groups, and a separate learning model is used to predict flight delays for each group. This strategy allows the problem to be decomposed into smaller sub-problems, leading to improved prediction system performance in terms of accuracy (by 2.49%) and processing speed (by 39.17%). The learning model used in each cluster is a novel structure based on a random forest, where each tree component is optimized and weighted using the Coyote Optimization Algorithm (COA). Optimizing the structure of each tree component and assigning weighted values to them results in a minimum 5.3% increase in accuracy compared to the conventional random forest model. The performance of the proposed method in predicting flight delays is tested and compared with previous research. The findings demonstrate that the proposed approach achieves an average accuracy of 97.2% which indicates a 4.7% improvement compared to previous efforts.

Evaluation of the Application Effect of Enteral and Parenteral Nutrition Therapy Combined with a Health Belief Education Model in Patients with Inflammatory Bowel Disease

Objective: To evaluate the application effect of enteral and parenteral nutrition therapy combined with a health belief education model in patients with inflammatory bowel disease. Methods: 80 patients with inflammatory bowel disease admitted to the Shanghai Zhangjiang Institute of Medical Innovation were chosen. This study was carried out from August 2022 to October 2023. The patients were randomly divided into a study group (40 cases) and a control group (40 cases). The treatment plan for the control group was the conventional treatment model, while the treatment plan for the study group was to provide enteral and parenteral nutrition therapy combined with a health belief education model based on the control group. The efficacy of both groups was compared. Results: In the study group, the therapeutic effect for 31 patients (77.50%) was markedly effective and 7 was effective (17.50%), accounting for 95.0% of the total, which was higher than the control group at 80.0% (P < 0.05). The relief time of relevant symptoms in the study group was shorter than that of the control group (P < 0.05). Before treatment, there were no differences in the high-sensitivity C-reactive protein (hs-CRP), interleukin 10 (IL-10), and tumor necrosis factor- α (TNF-α) between both groups (P > 0.05). After treatment, the levels of inflammatory factors in the study group (hs-CRP (8.02 ± 1.13) mg/L, IL -10 (9.24 ± 1.25) pg/mL, and TNF-α (7.19 ± 1.04) ng/L) were lower than those in the control group (P < 0.05). Conclusion: Enteral and parenteral nutritional therapy combined with a health belief education model showed significant efficacy in inflammatory bowel disease patients. Patient symptoms were relieved and inflammatory reactions were reduced. This method is worthy of popularization.

Exploring representative samples for modeling of wave buoy motion behavior

To collect sufficient data to describe buoy motion behavior in extreme environment conditions is intractable. The conventional data-driven models constructed with randomly selected and limited training samples are unable to perform adequately for exhibiting complex dynamic characteristics. In this work, an active sample exploring strategy is developed to build a reliable prediction model with limited samples. First, an evaluated index is designed to depict both the data distribution characteristics and estimation uncertainty of all test data using a probabilistic model. The representative data are then captured and integrated into the initial training set based on this index, and the updated uncertainty information is used as the criterion to judge the exploring termination. Finally, the initial training set is supplemented to enhance the model performance. Experimental and comparative results demonstrate the superiority of the proposed sample exploring strategy.

Numerical treatment for double diffusion phenomenon with generalized heat flux effects in 3d nanomaterial flow over bi-directional stretched wall

Current study aims to numerically investigate the bi-directional flow of nanofluids in the presence of Cattaneo–Christov double diffusion for two heat sources, namely, prescribed surface temperature (PST) along with prescribed heat flux (PHF). The conventional mathematical model in partial differential equations (PDEs) of the system have been reduced into an equivalent set of ordinary differential equations (ODEs). The system dynamics in the form of approximate solutions of the ODEs is presented by Adams and Explicit Runge Kutta numerical solvers. For better understanding the influence of physical parameter of interest including Brownian movement parameter, thermophoresis parameter, concentration relaxation parameter, Schmidt number, Prandtl number, thermal relaxation parameter, temperature ratio parameter, stretching ratio parameter and Deborah number on temperature, as well as concentration profiles are presented exhaustively. Appropriateness of the scheme is ascertained through close resemblance of results for different state of the art counterparts with negligible error around 10–05 to 10–09.

A comprehensive high‐level automated driving assistance system with integrated multi‐functionality

AbstractAdvanced Driver Assistance Systems (ADAS) have gained substantial attention in recent years. However, the integration mechanism of multiple functions within ADAS remains unexplored, and the full potential of its functionality remains underutilised. This paper presents a novel multi‐functional integrated High‐level Automated Driving Assistance System that combines the Cruise Control (CC), Adaptive Cruise Control (ACC), Automated Emergency Brake (AEB), and Automated Lane Change (ALC) functions. The presented system utilises a hierarchical framework. The extension multi‐mode switch strategy is established as the superior module and the Event‐Triggered Model Predictive Controller (ETMPC) is designed as the inferior controller. The CC, ACC, and ALC functions are effectively utilised to enhance traffic efficiency, while the AEB function ensures driving safety. To address the time constraints of conventional Model Predictive Control, an event‐trigger mechanism is proposed to reduce computational load. Simulations are conducted using the CarSim and Matlab platforms. The study results demonstrate significant improvements in both safety and traffic efficiency compared to conventional ADAS strategies. Furthermore, the proposed ETMPC method significantly reduces the frequency of solving Optimisation Problems and decreases online computation costs.

A composite water saturation model of continental mixed shale oil reservoirs based on complex lithology identification

The complex lithology, frequent interbedding and strong heterogeneity of the shale oil reservoir in Lucaogou Formation of the Middle Permian in Jimsar Sag lead to the complexity of litho‐electric parameters and the low accuracy of conventional reservoir saturation evaluation methods. Therefore, in this paper, automatic lithologic identification was firstly performed with the optimization inversion method of mineral components and K‐means clustering algorithm, and then, litho‐electric experiments were carried out to determine the litho‐electric parameters of different lithologies. Furthermore, the shale oil reservoir saturation calculation models of different lithologies were optimized and determined. Then, a composite water saturation model (Composite‐SW) based on complex lithology identification was constructed. Finally, the model was verified and analysed with the experimental results of water saturation. The coefficient of correlation between the calculation results of the Composite‐SW model and the core experiment results reached .88, indicating that the model was much better than conventional models. The model can be widely applied in logging evaluation of the continental mixed shale oil reservoir in the study area. The study provides a new idea for oil saturation evaluation of continental shale oil reservoirs and is significant for evaluating other types of shale oil reservoirs.

Social determinants of health and diabetes: using a nationally representative sample to determine which social determinant of health model best predicts diabetes risk

ObjectivesSocial determinants of health (SDOH) research demonstrates poverty, access to healthcare, discrimination, and environmental factors influence health outcomes. Several models are commonly used to assess SDOH, yet there is limited understanding of how these models differ regarding their ability to predict the influence of social determinants on diabetes risk. This study compares the utility of four SDOH models for predicting diabetes disparities.Study designWe utilized The National Longitudinal Study of Adolescent to Adulthood (Add Health) to compare SDOH models and their ability to predict risk of diabetes and obesity.MethodsPrevious literature has identified the World Health Organization (WHO), Healthy People, County Health Rankings, and Kaiser Family Foundation as the conventional SDOH models. We used these models to operationalize SDOH using the Add Health dataset. Add Health data were used to perform logistic regressions for HbA1c and linear regressions for body mass index (BMI).ResultsThe Kaiser model accounted for the largest proportion of variance (19%) in BMI. Race/ethnicity was a consistent factor predicting BMI across models. Regarding HbA1c, the Kaiser model also accounted for the largest proportion of variance (17%). Race/ethnicity and wealth was a consistent factor predicting HbA1c across models.ConclusionPolicy and practice interventions should consider these factors when screening for and addressing the effects of SDOH on diabetes risk. Specific SDOH models can be constructed for diabetes based on which determinants have the largest predictive value.

Event-Triggered Relearning Modeling Method for Stochastic System with Non-Stationary Variable Operating Conditions

This study presents a novel event-triggered relearning framework for neural network modeling, designed to improve prediction precision in dynamic stochastic complex industrial systems under non-stationary and variable conditions. Firstly, a sliding window algorithm combined with entropy is applied to divide the input and output datasets across different operational conditions, establishing clear data boundaries. Following this, the prediction errors derived from the neural network under different operational states are harnessed to define a set of event-triggered relearning criteria. Once these conditions are triggered, the relevant dataset is used to recalibrate the model to the specific operational condition and predict the data under this operating condition. When the predicted data fall within the training input range of a pre-trained model, we switch to that model for immediate prediction. Compared with the conventional BP neural network model and random vector functional-link network, the proposed model can produce a better estimation accuracy and reduce computation costs. Finally, the effectiveness of our proposed method is validated through numerical simulation tests using nonlinear Hammerstein models with Gaussian noise, reflecting complex stochastic industrial processes.

THEORETICAL AND NUMERICAL ANALYSIS OF FUEL DROPLET COMBUSTION PROCESS WITH CONVENTIONAL AND TRANSIENT MODELS

The study introduces a theoretical analysis and numerical solution for the combustion of a hydrocarbon fuel droplet inside the combustion chamber. The study employs two mathematical models to analyze the combustion process. Conventional (Classical) model and Transient model. The combustion process of a stagnant droplet in the steady state was analyzed in the classical model, while, In the transient model, it was assumed that there is a period of time in the stages of the droplet combustion in which the droplet is heated before combustion. The effect of change in temperature on the thermo-physical properties of the fuel was adopted through the two models. For the classical model, a convenient approximation was adopted for the heat transferred inside the droplet.

Intensification of daily tropical precipitation extremes from more organized convection.

Tropical precipitation extremes and their changes with surface warming are investigated using global storm resolving simulations and high-resolution observations. The simulations demonstrate that the mesoscale organization of convection, a process that cannot be physically represented by conventional global climate models, is important for the variations of tropical daily accumulated precipitation extremes. In both the simulations and observations, daily precipitation extremes increase in a more organized state, in association with larger, but less frequent, storms. Repeating the simulations for a warmer climate results in a robust increase in monthly-mean daily precipitation extremes. Higher precipitation percentiles have a greater sensitivity to convective organization, which is predicted to increase with warming. Without changes in organization, the strongest daily precipitation extremes over the tropical oceans increase at a rate close to Clausius-Clapeyron (CC) scaling. Thus, in a future warmer state with increased organization, the strongest daily precipitation extremes over oceans increase at a faster rate than CC scaling.

Lightweight small target detection based on aerial remote sensing images

With the upgrading of aviation space technology, the amount of information contained in remote sensing images in the aviation is gradually increasing, and the detection technology based on small targets has developed. For lightweight small targets, pixels per unit area contain more information than large targets, and their area is too small, which is easily overlooked by conventional detection models. To enhance the attention of such algorithms, this study first introduces a Control Bus Attention Mechanism (CBAM) in the fifth generation You Only Look Once (YOLOv5) algorithm to increase the algorithm’s attention to small targets and generate optimization algorithms. Then convolutional neural network is used to mark feature pixels of the optimization algorithm, eliminate redundant information, and generate fusion algorithm, which is used to generate redundant information with high similarity when the optimization algorithm surveys pixel blocks. The novelty of this study lies in using CBAM to improve YOLOv5 algorithm. CBAM module can extract important features from images by adaptively learning the channel and spatial attention of feature maps. By weighting the channel and spatial attention of the feature map, the network can pay more attention to important features and suppress irrelevant background information. This attention mechanism can help the network better capture the characteristics of small targets and improve the accuracy and robustness of detection. Embedding CBAM module into YOLOv5 detection network can enhance the network's perception of small targets. CBAM module can improve the expressive ability and feature extraction ability of the network without increasing the complexity of the network. By introducing CBAM module, YOLOv5 can better capture the characteristics of small targets in aerial remote sensing images, and improve the detection accuracy and recall rate. Finally, the proposed fusion algorithm is used for experiments on the Tiny-Person dataset and compared with the fifth, sixth, and seventh generations of You Only Look Once. When the fusion algorithm tests the target, the classification accuracy of Sea-person is 39 %, the classification accuracy of Earth-person is 31 %, and the probability of being predicted as the background is 56 % and 67 %, respectively. And the overall accuracy of this algorithm is 0.987, which is the best among the four algorithms. The experimental results show that the fusion algorithm proposed in the study has precise positioning for lightweight small targets and can achieve good application results in aerial remote sensing images.

Efficient super-resolution image reconstruction of electrical capacitance tomography for gas–solid fluidized bed measurement

Electrical capacitance tomography (ECT) has been widely used to investigate the flow dynamics of the gas–solid fluidized bed. Deep learning-based tomographic imaging of ECT is of great potential to reconstruct the particle concentration distribution with compatible image quality and reconstruction speed. However, it suffers from low model construction efficiency as numerous data samples need to be collected for training the deep neural network. Here a super-resolution (SR) imaging model based on a two-stage imaging network is proposed to improve the model construction efficiency. A low-resolution (LR) imaging network is used to reconstruct the LR image with the measured capacitance matrix while a subsequent high-resolution (HR) network calculates the HR image with the reconstructed LR image. The imaging performance, construction efficiency, and generalization ability of the trained SR imaging model are evaluated by reconstructing the samples in the test set and the unseen typical flow patterns. Considering the training set includes 100,000 data samples, the time consumption of numerically collecting the dataset for training the SR imaging model is 2.70 h, 2.2 % of that for a conventional deep learning-based imaging model of ECT, and the overall model construction time is only 4.44 h when the grid numbers of the LR and HR images are 246 and 8053, respectively. Experimental measurements are further carried out on a fluidized bed and the preliminary results show good prediction and generality of the trained model for reconstructing the particle concentration distribution.

NIR spectroscopy-CNN-enabled chemometrics for multianalyte monitoring in microbial fermentation.

As the biopharmaceutical industry looks to implement Industry 4.0, the need for rapid and robust analytical characterization of analytes has become a pressing priority. Spectroscopic tools, like near-infrared (NIR) spectroscopy, are finding increasing use for real-time quantitative analysis. Yet detection of multiple low-concentration analytes in microbial and mammalian cell cultures remains an ongoing challenge, requiring the selection of carefully calibrated, resilient chemometrics for each analyte. The convolutional neural network (CNN) is a puissant tool for processing complex data and making it a potential approach for automatic multivariate spectral processing. This work proposes an inception module-based two-dimensional (2D) CNN approach (I-CNN) for calibrating multiple analytes using NIR spectral data. The I-CNN model, coupled with orthogonal partial least squares (PLS) preprocessing, converts the NIR spectral data into a 2D data matrix, after which the critical features are extracted, leading to model development for multiple analytes. Escherichia coli fermentation broth was taken as a case study, where calibration models were developed for 23 analytes, including 20 amino acids, glucose, lactose, and acetate. The I-CNN model result statistics depicted an average R2 values of prediction 0.90, external validation data set 0.86 and significantly lower root mean square error of prediction values ∼0.52 compared to conventional regression models like PLS. Preprocessing steps were applied to I-CNN models to evaluate any augmentation in prediction performance. Finally, the model reliability was assessed via real-time process monitoring and comparison with offline analytics. The proposed I-CNN method is systematic and novel in extracting distinctive spectral features from a multianalyte bioprocess data set and could be adapted to other complex cell culture systems requiring rapid quantification using spectroscopy.

Development of single point prediction model using artificial neural network and experimental validation for pump as turbine applications

This paper proposed a new approach based on artificial neural networks (ANNs) to select the most appropriate centrifugal pumps for reverse operation i.e. pump as turbines (PAT), by predicting parameters at Best Efficiency Point from pump mode data. Both, exhaustive experimental data of 49 pumps (specific speeds 9−104) collected from open literature and by in-house experimentation data, considered input data-set as pump and target data-set as PAT are used for training the ANN models with two different training functions: Levenberg–Marquardt and Bayesian regularisation. The proposed ANN prediction model shows a deviation lower than 10% compared to the respective experimental value. Additionally, the trained ANN model tested with four pumps (not included in training data sets of proposed models) shows maximum absolute deviation in head number, flow number, and efficiency within the 10%, 7%, and 6% range, respectively, compared to the experimental values. Furthermore, the efficiency of reverse mode evaluated for variation in rotational speed shows a maximum of 3.3% (within 5%) absolute deviation compared to respective experimental results. Overall, the ANNs based prediction model of PAT parameters is recommended compared to conventional models available in the literature as it gives superior results (less than 10% deviation) for practical application.

TS-Fastformer: Fast Transformer for Time-series Forecasting

Many real-world applications require precise and fast time-series forecasting. Recent trends in time-series forecasting models are shifting from LSTM-based models to Transformer-based models. However, the Transformer-based model has a limited ability to represent sequential relationships in time-series data. In addition, the transformer-based model suffers from slow training and inference speed due to the bottleneck incurred by a deep encoder and step-by-step decoder inference. To address these problems, we propose a time-series forecasting optimized Transformer model, called TS-Fastformer. TS-Fastformer introduces three new optimizations: First, we propose a Sub Window Tokenizer for compressing input in a simple manner. The Sub Window Tokenizer reduces the length of input sequences to mitigate the complexity of self-attention and enables both single and multi-sequence learning. Second, we propose Time-series Pre-trained Encoder to extract effective representations through pre-training. This optimization enables TS-Fastformer to capture both seasonal and trend representations as well as to mitigate bottlenecks of conventional transformer models. Third, we propose the Past Attention Decoder to forecast target by incorporating past long short-term dependency patterns. Furthermore, Past Attention Decoder achieves high performance improvement by removing a trend distribution that changes over a long period. We evaluate the efficiency of our model with extensive experiments using seven real-world datasets and compare our model to six representative time-series forecasting approaches. The results show that the proposed TS-Fastformer reduces MSE by 10.1% compared to state-of-the-art model and demonstrates 21.6% faster training time compared to the existing fastest transformer, respectively.

Differential evolution–based integrated model for predicting concrete slumps

Concrete slump, a crucial indicator of fluidity, directly affects the pumpability and construction efficiency of concrete. Conventionally, concrete slumps are assessed through multiple test iterations conducted by skilled professionals to obtain accurate results. In this study, a predictive model was established to predict concrete slumps directly based on concrete mix proportions, thereby mitigating labor and time costs. An open-source dataset was used in this study. The selected water-cement ratio ranged from 0.29 to 0.66. Cement, slag, fly ash, water, superplasticizer, coarse aggregate, and fine aggregate were employed as input parameters. Furthermore, the performances of a variety of algorithms in predicting concrete slumps were analyzed and compared. The algorithms included SVM, KNN, Extra-Trees, Gradient Boosting, Decision Tree, Elastic Net, Lasso, Ridge, Random Forest, Bagging, AdaBoost, and XGBoost. Through a comprehensive comparison of the prediction performance of these models, AdaBoost, Bagging, Extra-Trees, and Random Forest were adopted as base models. Moreover, the SVM algorithm optimized using Differential Evolution was employed as a secondary model to construct an enhanced integrated prediction model. The final prediction model boasts an MSE of 2.099984, MAE of 1.225597, RMSE of 1.449132, and R2 of 0.970418. Compared to conventional prediction models, the proposed model has the potential to substantially improve prediction performance.