Search

The present study undertakes a comprehensive assessment of two predictive models, namely Long Short-Term Memory (LSTM) and Multi-layer Perceptron (MLP), with a specific emphasis on their effectiveness in predicting oil prices, particularly those of the Petroleum Exporting Countries (OPEC). In this study, three fundamental statistical measures are utilized: The Symmetric Mean Absolute Percentage Error (SMAPE), the Mean Squared Error (MSE), and the Mean Absolute Percentage Error (MAPE). The results demonstrate that the LSTM model regularly surpasses the MLP model in the three benchmarks. In particular, the LSTM model demonstrates lower values for SMAPE, MSE, and MAPE, indicating higher prediction accuracy. The decreased error scores linked to the LSTM model highlight its improved capacity for precise oil price prediction in comparison to the MLP model. These results signify a notable progress in the use of machine learning techniques for predicting OPEC oil prices. Moreover, this study provides invaluable perspectives for OPEC management, policymakers, and organizations focused on oil price fluctuations, therefore contributing to the wider endeavour of enhancing the stability and economic sustainability of the oil pricing system in OPEC countries. The consequences of the study include the promotion of a pricing system that facilitates the achievement of economic and social development goals in these countries.

Speaking of speech recognition within the English language, it is the process of recognizing oral speech and transcribing it into writing using exclusive algorithms. For the perishable skill of English language learning, use of innovative speech recognition technology using Advanced Speech Recognition Technologies MLP-LSTM is proposed in this paper to advance the existing online learning platforms. Previous research addresses the importance of NLP in English language learning but notes the challenges in effectively extracting and segmenting features from multimodal data. In order to overcome these problems, this paper incorporate the proposed MLP for feature extraction and LSTM for sequence learning. The utilization of MLP-LSTM provides not only a brilliant improvement of the capacity to transform spoken language and perceive it but also minimizes the Word Error Rate (WER) to 0.075. With this low WER, along with the total accuracy rate of 98.25 %, this paper focus on underlining how this system is more effective than traditional language learning tools. This paper has been implemented through Python Software. The given MLP-LSTM based speech recognition model lays the foundation for a highly complex yet accurate paced English language learning platform that will cater to the needs of the learners in the global scenario.

Solar energy is one of the renewable and clean energy sources. Accurate solar radiation (SR) estimates are therefore needed in solar energy applications. Firstly, two deep learning models, including gated recurrent unit (GRU) and long short-term memory (LSTM), were developed in this study. Next, a data pre-processing technique named multivariate variational mode decomposition (MVMD) was used to construct the MVMD-GRU and MVMD-LSTM hybrid models. To better test the performance of proposed simple and hybrid models, four stations located in the Illinois State of the USA (i.e., Dixon Springs, Fairfield, Rend Lake, and Carbondale) were considered as the study sites. Whole the simple and hybrid models were established under two different strategies, i.e., local and external. In the local strategy, SR of each location was estimated using the minimum and maximum air temperatures from the same station. While, minimum and maximum air temperatures as well as SR data from the nearby station were utilized in external strategy to estimate SR time series of any target site. Root mean square error (RMSE), mean absolute error (MAE), and coefficient of determination (R2) metrics were used when evaluating the models performances. The overall results revealed that the proposed MVMD-GRU and MVMD-LSTM hybrid models illustrated better SR estimates compared to the simple GRU and LSTM in both the local and external strategies. The values of error metrics obtained for the superior hybrid models (i.e., MVMD-LSTM) during the testing period were as: RMSE = 2.532 MJ/m2.day, MAE = 1.921 MJ/m2.day, R2 = 0.916 at Dixon Springs; RMSE = 2.476 MJ/m2.day, MAE = 1.878 MJ/m2.day, R2 = 0.921 at Fairfield; RMSE = 2.359 MJ/m2.day, MAE = 1.780 MJ/m2.day, R2 = 0.924 at Rend Lake; RMSE = 2.576 MJ/m2.day, MAE = 1.941 MJ/m2.day, R2 = 0.914 at Carbondale. Therefore, the coupled models proposed in this study can be possibly recommended as suitable alternatives to the simple deep learning models with a reliable precision in estimating SR time series.

State of charge (SOC) plays a vital role in the safe, efficient, and stable operation of lithium-ion batteries. Since the difference between the surface temperature and core temperature of batteries under severe conditions can reach 5–10 ​°C, using the surface temperature as input feature of SOC estimation is unreasonable. Due to the high requirement for storage space, SOC estimation methods based on deep learning methods are limited to implement in embedded devices. In this paper, to achieve reasonable and high accuracy SOC estimation and provide support for battery thermal management, SOC estimation based on state of temperature (SOT) is implemented. And weight clustered-convolutional neural network-long short-term memory (WC-CNN-LSTM) is proposed to achieve high accuracy SOT and SOC estimation with small model sizes. A self-established dataset is used to verify the effectiveness of the proposed method and model. The WC-CNN-LSTM model with the number of clusters of 400 could achieve comparative accuracy with the baseline model with a 52.98% smaller model size and 25.08% more time consumption for model training on SOT estimation. And it could also achieve consistent and even better accuracy on SOC estimation with the baseline model with a small model size.

Global adoption of wind energy continues to increase, while improving the efficiency of turbine settings requires reliable wind speed (WS) models. The latest models rely on artificial intelligence (AI) optimizations which constructs tests on a range of novel hybrid models to examine the reliability. Gradient Boosting (GB), Random Forest (RF), and Long Short-Term Memory (LSTM) are used in new combinations for data pre-processing. A Time Varying Filter-based Empirical Mode Decomposition (TVFEMD) model is coupled with the GB and LSTM standalone models, to create TVFEMD-GB and TVFEMD-LSTM hybrids, which are run in competition with each other. Eventually, a preferred hybrid form is established, simultaneous hybridization of TVFEMD with GB and LSTM. This study is the first to hybridize these fundamental systems, and create a TVFEMD-GB-LSTM model that can forecast WS. This study finds that the novel hybrid models exhibit superior performance to standalone GB and LSTM models, opening the pathway to alternative WS prediction techniques.

To ensure the safe operation of lithium-ion batteries, it is crucial to accurately predict their state of health (SOH) and remaining useful life (RUL). Addressing the issue of high costs and time consumption due to the reliance on large amounts of labeled data in existing models, this paper proposes a co-estimation framework that combines semi-supervised learning (SSL) with long short-term memory networks (LSTM), effectively utilizing unlabeled data. By selecting the most strongly correlated battery health features and constructing a degradation model using a hybrid dataset, the need for labeling is reduced. The verification results indicate SOH estimated error is reduced to 4 % and the maximum root mean square error (RMSE) is 1.58 %. When utilizing 75 % SOH as the end-of-life criterion for battery cycle life, the mean absolute error (MAE) of the RUL predictions for the two tested batteries are 2.5281 and 0.0562 cycles, respectively. The results prove the framework enables accurate prediction and has wide practicability and universal applicability.

In our study, we introduce a novel hybrid ensemble model that synergistically combines LSTM, BiLSTM, CNN, GRU, and GloVe embeddings for the classification of gene mutations in cancer. This model was rigorously tested using Kaggle's Personalized Medicine: Redefining Cancer Treatment dataset, demonstrating exceptional performance across all evaluation metrics. Notably, our approach achieved a training accuracy of 80.6 %, precision of 81.6 %, recall of 80.6 %, and an F1 score of 83.1 %, alongside a significantly reduced Mean Squared Error (MSE) of 2.596. These results surpass those of advanced transformer models and their ensembles, showcasing our model's superior capability in handling the complexities of gene mutation classification. The accuracy and efficiency of gene mutation classification are paramount in the era of precision medicine, where tailored treatment plans based on individual genetic profiles can dramatically improve patient outcomes and save lives. Our model's remarkable performance highlights its potential in enhancing the precision of cancer diagnoses and treatments, thereby contributing significantly to the advancement of personalized healthcare.

Wind power forecasting is crucial for energy conversion and management. This study employs the long short-term memory (LSTM) network, a specialized form of recurrent neural networks (RNNs) noted for its effectiveness in time-series prediction, to predict wind power from various turbines. Furthermore, we incorporate a cutting-edge metaheuristic optimization technique to optimize the training process of the LSTM, enhancing its parameter optimization. Specifically, we utilize the attraction–repulsion optimization algorithm (AROA), an innovative optimization algorithm inspired by the natural phenomena of attraction and repulsion for addressing complex optimization and engineering problems. This research applies the AROA to optimize the LSTM training process, which markedly improves the model's forecasting accuracy. Our analysis is conducted using four datasets from La Haute Borne wind turbines in France. The proposed AROA-LSTM model achieved R2 testing results of 0.9416, 0.9663, 0.9613, and 0.9622 for Turbines 1, 2, 3, and 4, respectively.

Accurate hydrological predictions are often hindered by the lack of stream gauges in data-scarce regions, where traditional transfer learning (TL) models like Long Short-Term Memory (LSTM) networks often face limitations due to reduced accuracy and adaptability. To enhance runoff prediction in such regions, we developed DAformer, a novel TL approach that integrates domain adversarial neural networks with the Informer model. Trained on comprehensive runoff data from U.S. basins, DAformer was applied to three basins in Chile and the Chaersen basin in China, demonstrating an effective transfer from data-rich to data-scarce environments. Results show that DAformer significantly outperforms LSTM-based models, improving forecast accuracy by 16.1% for 1-day lead time and by 100.5% for 5-day lead time. These improvements indicate that the DAformer model not only enhances prediction accuracy but also holds substantial practical implications for flood risk management and water resource planning in regions with limited data availability. By clustering basins based on Shuttle Radar Topography Mission (SRTM) and other geographical data, we found that relying on multiple source basins further enhances the performance. DAformer, therefore, serves as a robust and scalable method for enhancing runoff prediction for regions with limited data.

Urban traffic congestion presents a range of vital difficulties requiring precise prediction models in order to facilitate traffic management for Autonomous Vehicles. This work introduces a novel framework that regulates a Long Short-Term Memory (LSTM) system with methods provided by Explainable Artificial Intelligence (XAI) to explain traffic congestion behavioural modes. For enhanced accuracy and transparency, the integration of EAI methodologies with LSTM based models is addressed as a novel approach towards congestion prediction, while significant research has been done previously using Machine Learning that compared previous proposed based model congestion monitoring improvement through Federated Learning Waqas et al. [18]. This wok proposes the enhances ML focused on Long Short-Term Memory with EAI (LSTM-EAI) model for Smart City environments that require accurate traffic congestion rate forecast to improve the urban mobility. The proposed model provides better interpretability that help stakeholders to understand how the input plays an important role in the condition of traffic jams. The results show that the LSTM-EAI model is 5 % better than previous methods for both the accuracy and reliability of congestion prediction, and may become a practical and effective solution for the urban traffic problem.

Study region: The Soyang Reservoir and its upstream river located in a monsoon climate region in South Korea, where rapid turbidity changes significantly impact reservoir operations. Study focus: This study aimed to enhance turbidity predictions in stratified reservoirs and address the challenge of insufficient boundary condition data by developing a hybrid approach combining the CE-QUAL-W2 (W2) process-based model (PBM) with machine learning (ML) techniques. Various ML models were compared against conventional methods for generating turbidity boundary condition data. New hydrological insights for the region: The hybrid ML-PBM approach significantly improved turbidity predictions, especially during rapid changes associated with rainfall events in the Soyang Reservoir. The long short-term memory model outperformed other ML methods, increasing the W2 model efficiency index by 1.46-fold compared with traditional regression methods. This approach effectively captured complex nonlinear relationships in turbidity dynamics influenced by flow rate and precipitation. The hybrid ML-PBM model accurately predicted high-turbidity event durations crucial for managing downstream ecosystems and water treatment processes. Its potential for enhancing water quality modeling and reservoir management strategies in monsoon-affected regions worldwide was demonstrated, offering a robust solution for predicting water quality dynamics in reservoirs with limited inputs. The hybrid method's success in the Soyang Reservoir suggests its applicability to similar reservoirs globally, potentially improving water resource management in regions facing comparable challenges.

The growing demand for renewable energy sources like wind and solar power requires accurate and reliable forecasting techniques for effective planning and operation. This study presents an attention-based spatial-temporal graph neural network–long short-term memory (ASTGNN-LSTM) model designed to predict wind speed and solar radiation using 20 years of meteorological data from five regions in Northwest China. The ASTGNN-LSTM model shows significant performance improvements over traditional methods, such as the historical average model, autoregressive integrated moving average model, and graph convolutional network with LSTM. After optimizing the hidden layers and learning rate, the relative errors for predicting wind speed and solar radiation are reduced to 27.15 % and 6.11 %, respectively. Sensitivity analysis reveals that location data have the most significant impact on predictions. These findings demonstrate that the ASTGNN-LSTM model effectively captures nonlinear relationships and can enhance renewable energy planning and management.

In response to the problems of insufficient model accuracy and poor real-time performance in real-time monitoring and optimization of data models in traditional power systems, this paper explored them based on machine learning. It used methods such as box plots and Kalman filters to preprocess power data and used Fourier transform and long short-term memory (LSTM) network to extract data features. Based on long-term and short-term memory networks, this paper constructed a deep neural network model to process data. Recursive feature elimination method can be used for feature selection in the model, and multi-model integration method can be used for feature fusion. The experiment trained and tested the model on two datasets, IEEE 39-Bus and Pecan Street. The experimental results show that the accuracy of the paper's model in both the training and testing sets is higher than that of the convolutional neural network, decision tree, and support vector machine models in the comparative experiments, reaching 93 % and 94 %, respectively. The average response times in three different testing methods were 139.8 ms, 151 ms, and 140.6 ms, respectively. The lowest daily failure rate for 30 consecutive days of work was 0 %, which reflects the stability of the model; in the satisfaction survey of practitioners, the proportion of high recognition answers to each question was higher than 80 %. The experimental results show that the machine learning intelligent control system has achieved good application in power data modeling technology, providing a reference for similar research in the future.

Energy efficiency in industrial systems remains a critical challenge, with traditional data-driven models often limited by model accuracy and data availability. Incorporation of physical laws governing energy systems can improve performance and physical consistency, but the model often struggles with the calculation of loss and ignores dynamic interplays between sub-systems, which can result in oversimplification and a lack of practical applicability. Therefore, this study investigated a theoretical framework for developing a law of conservation-guided neural network aimed at enhancing energy efficiency prediction in industrial systems. The framework integrates physical principles directly into floating nodes constructed using a long short-term memory architecture to help the model formulate the relationship between process variables, while gradient aggregation increases liquidity and interpretability. Through evaluation of two large-scale case studies—vinyl chloride monomer and detergent powder production—the proposed model produced substantial improvements in prediction accuracy and model reliability, with a test prediction improvement of 12.2 % and 5.87 % over published methods. Compared to network architecture modification approaches, the proposed model provided higher reliability and reproducibility in energy efficiency predictions. Moreover, the model successfully identified energy inefficiencies, resulting in a 4.21 % reduction in energy consumption and a corresponding 377.35 tons of carbon emissions reduction.

In order to enhance grid stability, facilitate efficient power supply planning, and minimize energy wastage, precise prediction of building energy usage stands as a significant worldwide issue. This paper proposes a prediction model based on multi-layer Long Short-Term Memory (LSTM) coupled with data denoising techniques, including the continuous wavelet transform (CWT) and Kalman filter (KF) to attain superior prediction performance. The primary goal of KF and CWT for time series data is denoising the data to enhance the signal and remove noise, making it easier to identify and isolate relevant patterns. The results from a comparative study focusing on urban household electricity consumption in Morocco with different step sizes, along with five extended applications, demonstrate that the proposed model outperforms basic LSTM and other existing models. The Mean Absolute Percentage Error (MAPE) for the comparative study with different step sizes stands at 4.02 % and 2.95 % respectively. Additionally, the MAPE for the five extended applications across different buildings are 0.88 %, 3.89 %, 0.69 %, 0.89 %, 0.87 %, respectively. These results highlight the superior accuracy of the proposed model across all metrics, making it a promising technique for predicting building energy consumption. This accuracy is paramount for optimizing energy management strategies, enabling informed decision-making and resource allocation in pursuit of sustainable and efficient building operations.

The compressive strength of ultrahigh performance concrete (UHPC) is influenced by the composition, quality, and quantity of its constituent elements. Using traditional statistical methods, it is difficult for us to quantify the relationships between the technical properties of UHPC and the composition of the mixture because of their complexity and nonlinearity. This work aims to develop advanced prediction models for estimating UHPC compressive strength over a large spectrum of supplementary cementitious material combinations and aggregate sizes. The models trained on the UHPC mixture dataset with 15 input variables included the group method of data handling, recurrent neural networks, long short-term memory, and bidirectional long short-term memory (Bi-LSTM). These models routinely forecast UHPC compressive strength according to sensitivity analysis, external validation, and statistical performance measures. During testing, the Bi-LSTM model outperformed the other models, with an RMSE of 0.0482 and an R² value of 0.9464. These results maximize component selection by showing how effectively the Bi-LSTM model might reduce UHPC formulation development and lower the cost and testing time span.

Machine learning (ML) is emerging as a promising tool for modeling hydro-ecological processes due to the increasing availability of large environmental data. However, the use of ML requires sufficient programming knowledge due to a lack of a graphical user interface (GUI). In this study, we introduced a GUI package, named HydroEcoLSTM, with the long short-term memory network (LSTM) as the core model, that allows non-ML experts to utilize their domain knowledge to construct complex ML models. We demonstrated the functionalities of HydroEcoLSTM with two practical examples, including (1) predictions of streamflow in both gauged and ungauged catchments and (2) predictions of multiple outputs (i.e., streamflow and isotope transport from two catchments). The simulation results obtained in both case experiments are satisfactory. In the first example, the average Nash–Sutcliffe Efficiency (NSE) for streamflow simulation during the testing period is 0.79 while the application of the trained model in two assumed ungauged catchments also achieves the average NSE of 0.68. In the second example, the average NSE for streamflow and instream isotope simulation during the testing period is 0.71. Ultimately, applications of HydroEcoLSTM with real-world examples demonstrate its potential use for practical applications and research without requiring extensive coding skills.

The increasing occurrence of extreme weather events has severely compromised the resilience of power distribution systems, resulting in widespread outages and substantial economic losses. This paper proposes a novel solution to enhance the resilience of distribution networks without the need for significant infrastructure upgrades. We introduce a bilevel optimization framework that integrates Demand Response Programs (DRPs) to strategically manage electricity consumption and mitigate the impact of system disruptions. The approach fosters collaboration between Distribution System Operators (DSOs) and Demand Response Aggregators (DRAs), optimizing both operational resilience and economic efficiency. To solve the bilevel problem, we employ a Mathematical Program with Equilibrium Constraints (MPEC), transforming the bilevel model into a single-level problem by utilizing the Karush–Kuhn–Tucker (KKT) conditions. This method is applicable when the lower-level problem is convex with linear constraints. The model also incorporates Long Short-Term Memory (LSTM) neural networks for wind generation forecasting, enhancing decision-making precision. Furthermore, we conduct multiple case studies under varying severities of incidents to evaluate the method’s effectiveness. Simulations performed on the IEEE 33-bus test system using GAMS and Python validate that the proposed method not only improves system resilience but also encourages active consumer participation, making it a robust solution for modern smart grid applications. The simulation results show that by performing DRP to handle the contingencies in a high-impact incident, the resilience of the system can be improved by 5.3%.

Inflation growth in Indonesia and other countries impacts the currency value and investors' purchasing power, particularly in the transportation sector. This research explores the impact of inflation growth in Indonesia and comparable nations on currency valuation and the purchasing power of investors, with a focus on the transportation sector. Data collection was carried out from April to October 2023 by scraping stock data from several transportation stocks such as: AKSI.JK, CMPP.JK, SAFE.JK, SMDR.JK, TMAS.JK, and WEHA. The research primarily aims to forecast stock prices in Indonesia's transportation sector, utilizing data mining techniques within the Cross Industry Standard Process for Data Mining (CRISP-DM) framework, which includes stages such as business understanding, data preparation, modeling, evaluation, and deployment. It employs the Long Short-Term Memory (LSTM) algorithm, assessing different hyperparameter activation functions (linear, ReLU, sigmoid, tanh) and optimizers (ADAM, ADAGRAD, NADAM, RMSPROP, ADADELTA, SGD, ADAMAX) to refine prediction accuracy. Findings demonstrate the ReLU activation function and ADAM optimizer's effectiveness, highlighted by evaluation metrics such as Mean Absolute Error (MAE) of 0.0092918, Mean Absolute Percentage Error (MAPE) of 0.06422, and R-Squared of 96 %. The study notably identifies significant growth in Temas (TMAS.JK) stock from April to October 2023, surpassing other sector stocks. Additionally, a web-based application for predicting transportation stock prices has been developed, facilitating user inputs like ticker, activation-optimizer choice, and date range.

Background and Objective: Emerging diseases like Parkinson or Alzheimer's, which are not curable, endanger human mental health and are challenging to research. Drug target interactions (DTI) are pivotal in the screening of candidate drugs and focus on a small pool of drug targets. Electroencephalogram shows the responses to psychotropic medicines in the brain bioelectric activity. Synaptic activity can be analyzed by using Local Field Potential recordings obtained from micro-electrodes implanted in the brain. The aim is to evaluate the effects of drug on brain bioelectric activity and increase the drug classification accuracy. The ultimate goal is to advance our understanding of how drugs affect synaptic activity and open the door to more focused treatment for neurodegenerative diseases. Methods: In this study, Pharmaco-EEG recordings are processed using Advanced neural network models, particularly Convolutional Neural Networks, to assess the effects of medications. The five different medicines used in this study are Ephedrine, Fluoxetine, Kratom, Morphine, and Saline. The signals observed are local field potential signals. To overcome some limits of DTI prediction, we propose Bidirectional Long Short-Term Memory (LSTM) for the categorization of intracranial EEG (i-EEG) data, departing from standard approaches. Similar EEG patterns are presumably caused by drugs that work by homologous pharmacological pathways, producing similar psychotropic effects. To improve accuracy and reduce training loss, our study introduces a bidirectional LSTM model for classification along with Bayesian optimization Results: High recall, precision, and F1-Scores, particularly a 95% F1-Score for morphine, ephedrine, fluoxetine, and saline, suggest good performance in predicting these drug classes. Kratom produces a somewhat lower recall of 94%, but a high F1-Score of 97% and perfect precision of 1.00. The weighted average F1-Score, macro average, and overall accuracy are all consistently high (around 97%), indicating that the model works well throughout the spectrum of drugs. Conclusions: Improved model performance was demonstrated by using a diversified dataset with five drug categories and bidirectional LSTM boosted with Bayesian optimization for hyperparameter tuning. From earlier limited-category models, it represents a substantial advancement.

The growing energy demands of modern society have led to an increased reliance on secondary batteries, particularly lithium-ion (Li-ion) batteries, due to their superior energy density and power output. These batteries perform most effectively and safely within a specific temperature range, making it essential to develop accurate models for predicting temperature variations under diverse operational and environmental conditions. In particular, it is crucial to forecast temperature changes resulting from random and dynamic current fluctuations, reflecting real-world usage scenarios while considering the surrounding battery system environment. In this study, we employed a long short-term memory (LSTM) network to develop a surrogate model capable of predicting the battery’s core temperature over time, given varying current loads and heat transfer coefficients. The LSTM model demonstrated remarkable accuracy, achieving an average prediction accuracy of 99% in simulating temperature changes induced by arbitrary currents.

This paper aims to provide new avenues for innovation in credit governance in the digital economy to provide more reliable credit evaluation solutions for financial, commercial, and social interactions. This paper integrates the potential value of Internet of Things (IoT) technology in credit governance and proposes a credit governance method that utilizes IoT data and an improved Long Short-Term Memory model. The proposed model introduces an adaptive mechanism to monitor changes in data in real-time and automatically adjust network parameters to improve the model performance. Experimental validation is conducted on the University of California Irvine dataset. It is found that the proposed model performs well in terms of accuracy, F1 value, and Area Under the Curve value, with values of 0.9, 0.91, and 0.94, respectively. This indicates that the proposed model has higher classification accuracy and performance in classification tasks while maintaining efficient training time. This indicates that the presented method has potential application prospects in credit governance research.

Relative Oxygen Level of the Three-Way Catalyst is an important parameter that affects the conversion efficiency of pollutants. ROL is a time-varying hidden state variable that is difficult to directly observe in practice. Therefore, it is common to use a method of clearing oxygen storage to simplify control in vehicles. However, this method negates the positive effects of ROL on pollutant treatment. ROL can be indirectly observed through modeling methods. Chemical modeling methods involve extensive computational requirements that cannot meet the demands of practical control. In contrast, time-series neural networks offer computational speed advantages when dealing with similar problems. Therefore, the ROL observation models using both NARX and LSTM neural networks are developed and compared in this study. The results indicate that the NARX neural network exhibits higher precision with a smaller number of neurons and time steps. The LSTM neural network demonstrates greater stability when dealing with data error fluctuations. In practical applications, the ROL model can monitor the TWC operating status and assist in the development of intelligent pollutant aftertreatment control strategies.

Power forecasting plays a crucial role in the operation of smart grid system, which is indispensable for making the operation plan of power system, improving economic efficiency and ensuring the quality of power supply. In order to enhance the accuracy of power load forecasting, a hybrid intelligent power load forecasting system is proposed in this paper. The system first preprocesses the raw data using Savitzky-Golay smoothing technology to eliminate noise and improve data quality. Then, a long and short term memory network with attention mechanism is used to enhance the generalization ability of the model. In addition, in order to further improve the prediction performance, an improved genetic algorithm is integrated to optimize the model parameters. Finally, a data set is used to verify the proposed prediction method. In terms of short-term forecasting ability of experiment of the testing data set, compared with LSTM model, the proposed method shows superior performance in root mean square error and mean absolute error indicators, with root mean square error reduced by 18.7 % and mean absolute error reduced by 26.2 %. In terms of long-term prediction ability of experiment of the testing data set, compared with GBRT model, the proposed method reduces root mean square error and mean absolute error by 24.8 % and 30.7 %, respectively. The experimental results show that compared with the existing benchmark algorithm, the proposed method is significantly improved in two key indexes of prediction accuracy, which proves its effectiveness and superiority in power load prediction.

As critical transitional ecosystems, estuaries are facing the increasingly urgent threat of salt wedge intrusion, which impacts their ecological balance as well as human-dependent activities. Accurately predicting estuary salinity is essential for water resource management, ecosystem preservation, and for ensuring sustainable development along coastlines. In this study, we investigated the application of different machine learning and deep learning models to predict salinity levels within estuarine environments. Leveraging different techniques, including Random Forest, Least-Squares Boosting, Artificial Neural Network and Long Short-Term Memory networks, the aim was to enhance the predictive accuracy in order to better understand the complex interplay of factors influencing estuarine salinity dynamics. The Po River estuary (Po di Goro), which is one of the main hotspots of salt wedge intrusion, was selected as the study area. Comparative analyses of machine learning models with the state-of-the-art physics-based Estuary box model (EBM) and Hybrid-EBM models were conducted to assess model performances. The results highlighted an improvement in the machine learning performance, with a reduction in the RMSE (from 4.22 psu obtained by physics-based EBM to 2.80 psu obtained by LSBoost-Season) and an increase in the R2 score (from 0.67 obtained by physics-based EBM to 0.85 by LSBoost-Season), computed on the test set. We also explored the impact of different variables and their contributions to the predictive capabilities of the models. Overall, this study demonstrates the feasibility and effectiveness of ML-based approaches for estimating salinity levels due to salt wedge intrusion within estuaries. The insights obtained from this study could significantly support smart management strategies, not only in the Po River estuary, but also in other location.

In the broad field of sensors and 3D information retrieval, bathymetric reconstruction from side-scan sonar imaging is associated with unique technical hurdles. Neural Networks have recently led to promising new solutions in this field, but the available methods tend to be complex and data-intensive in a way typically making their use in a riverine environment impossible. Throughout our work, we have focused on simplifying the problem-handling and treating compatibility with a riverine environment as priority. In our work, Long Short-Term Memory proved to be effective in a surprisingly simple form. Combined with traditional post-processing techniques in the GIS environment, like median filtered focal statistics, our workflow ultimately results in ∼0.259 m median of error on the evaluation dataset of the Dráva River.

Installing photovoltaic (PV) systems in buildings is one of the most effective strategies for achieving sustainable energy goals and reducing carbon emissions. However, the requirement for efficient energy management, the fluctuating energy demands, and the intermittent nature of solar power are a few of the obstacles to the seamless integration of PV systems into buildings. These complexities surpass the capabilities of rule-based systems, necessitating innovative solutions. The research proposes a deep learning-based optimal energy management system designed specifically for home micro-grids that incorporate PV systems with battery energy storage, Enhanced Long Short-Term Memory (LSTM)-Based Optimal Home Micro-Grid Energy Management (OHM-GEM). Integrating an improved type of LSTM neural network called LSTM into the energy management system improves the reliability of PV power output predictions. The dependability of PV power production forecasts is increased by including a refined version of the LSTM neural network in the energy management system. The efficiency of the OHM-GEM system in maximizing PV system integration into buildings is shown by the authors using simulated data. With considerable gains in energy efficiency, cost savings, and decreased reliance on non-renewable energy sources, the results highlight the possibility of this approach to forward sustainable energy practices.

The precise assessment of shallow foundation settlement on cohesionless soils is a challenging geotechnical issue, primarily due to the significant uncertainties related to the factors influencing the settlement. This study aims to create an advanced hybrid machine learning methodology for accurately estimating shallow foundations' settlement (Sm). The initial contribution of the current research is developing and validating a robust hybrid optimization methodology based on an artificial electric field and single candidate optimizer (AEFSCO). This approach is thoroughly tested using various benchmark functions. AEFSCO will also be used to optimize three useful machine learning methods: long short-term memory (LSTM), support vector regression (SVR), and multilayer perceptron neural network (MLPNN) by adjusting their hyperparameters for predicting the settlement of shallow foundations. A database consisting of 189 individual case histories, conducted through various investigations, was used for training and testing the models. The database includes five input parameters and one output. These factors encompassed both the geometric characteristics of the foundation and the properties of the sandy soil. The results demonstrate that employing effective optimization strategies to adjust the ML models’ hyperparameters can significantly improve the accuracy of predicted results. The AEFSCO has increased the coefficient of determination (R2) value of the MLPNN model by 9.3 %, the SVR model by 8 %, and the LSTM model by 22 %. Also, the LSTM-AEFSCO model is more accurate than the SVR-AEFSCO and MLPNN-AEFSCO models. This is shown by the fact that R2 went from 0.9494 to 0.9290 to 0.9903, which is an increase of 4.5 % and 6 %.

Drowsiness is characterized by decreased alertness and an increased inclination to fall asleep, typically from factors such as fatigue, sleep deprivation, or other related influences. In the context of driving, drowsiness poses substantial safety risks. The detection of driver drowsiness is of paramount importance in ensuring road safety, contributing to a significant number of accidents worldwide. Utilizing AI for drowsiness detection offers a potent solution to enhance road safety by identifying driver fatigue and preventing potential accidents. The proposed system addresses the challenge of detecting driver drowsiness using a WACHSens dataset collected from both manual and automated driving modes, encompassing rested and fatigued states. Various data sources, including vehicle-related information, facial expressions, and bio signals are employed to create a robust drowsiness detection system. A novel approach that leverages Convolutional Neural Networks (CNNs) and Long Short-Term Memory (LSTM) networks to effectively detect drowsiness in drivers and achieve a 96 % accuracy level. It helps in enhancing road safety by devising effective drowsiness detection mechanisms, potentially preventing accidents and saving lives. Recall, accuracy, f1-score, and precision are the performance metrics to measure the drowsiness condition.

Compared with manual sorting of citrus fruit, vision-based sorting solutions can help achieve higher accuracy and efficiency. In this study, we present a vision system based on CNN-LSTM, which can cooperate with robotic grippers for real-time sorting and is readily applicable to various citrus processing plants. A CNN-based detector was adopted to detect the defective oranges in view and temporarily classify them into corresponding types, and an LSTM-based predictor was used to predict the position of the oranges in a future frame based on image sequential data. The fusion of CNN and LSTM networks enabled the system to track defective ones during rotation and identify their true types, and their future path was also predicted which is vital for predictive control of visually guided robotic grasping. High detection accuracy of 94.1% was obtained based on experimental results, and the error for path prediction was within 4.33 pixels 40 frames later. The average time to process a frame was between 28 and 62 frames per second, which also satisfied real-time performance. The results proved the potential of the proposed system for automated citrus sorting with good precision and efficiency, and it can be readily extended to other fruit crops featuring high versatility.

This paper presents an innovative fusion model called “CALSE-LSTM,” which integrates Convolutional Neural Networks (CNNs), Long Short-Term Memory Networks (LSTMs), self-attention mechanisms, and squeeze-and-excitation attention mechanisms to optimize the estimation accuracy of the State of Charge (SoC). The model incorporates battery historical data as input and employs a dual-attention mechanism based on CNN-LSTM to extract diverse features from the input data, thereby enhancing the model's ability to learn hidden information. To further improve model performance, we fine-tune the model parameters using the Pelican algorithm. Experiments conducted under Urban Dynamometer Driving Schedule (UDDS) conditions show that the CALSE-LSTM model achieves a Root Mean Squared Error (RMSE) of only 1.73 % in lithium battery SoC estimation, significantly better than GRU, LSTM, and CNN-LSTM models, reducing errors by 31.9 %, 31.3 %, and 15 %, respectively. Ablation experiments further confirm the effectiveness of the dual-attention mechanism and its potential to improve SoC estimation performance. Additionally, we validate the learning efficiency of CALSE-LSTM by comparing model training time with the number of iterations. Finally, in the comparative experiment with the Kalman filtering method, the model in this paper significantly improved its performance by incorporating power consumption as an additional feature input. This further verifies the accuracy of CALSE-LSTM in estimating the State of Charge (SoC) of lithium batteries.

The Internet of Things (IoT) landscape is witnessing rapid growth, driven by continuous innovation and a simultaneous increase in cybersecurity threats. As these threats become more sophisticated, the imperative to fortify IoT devices against emerging vulnerabilities becomes increasingly pronounced. This research is motivated by the need for comprehensive IoT threat detection solutions that can effectively address the evolving threat landscape. Existing approaches often fall short in adapting to the dynamic nature of IoT environments and the increasing complexity of attacks. The core problem addressed in this research is the development of a novel Hybrid Convolutional Neural Network and Long Short-Term Memory (CNN-LSTM) architecture tailored for precise and efficient IoT threat detection. This architecture aims to overcome the limitations of existing methods and enhance the security of IoT ecosystems. Our study encompasses a detailed analysis of the proposed Hybrid CNN-LSTM model, leveraging data from diverse datasets, including IoT-23, N-BaIoT, and CICIDS2017. The proposed model is tested and validated on more than 14 attack types. We have designed this model to exhibit robust threat detection capabilities by effectively capturing and analyzing IoT security data. The outcomes of our research showcase remarkable accuracy, with the models achieving 95% accuracy on the IoT-23 dataset and an outstanding 99% accuracy on both the N-BaIoT and CICIDS2017 datasets. These findings underscore the model's adaptability to various IoT environments. Our research contributes a comprehensive Hybrid CNN-LSTM architecture that significantly enhances IoT threat detection. We introduce Principal Component Analysis (PCA) to optimize data processing and incorporate advanced optimization techniques like model quantization and pruning to improve deployment efficiency in resource-constrained IoT environments. This study lays the foundation for future advancements in bolstering IoT security.

Water quality in surface bodies remains a pressing issue worldwide. While some regions have rich water quality data, less attention is given to areas that lack sufficient data. Therefore, it is crucial to explore novel ways of managing source-oriented surface water pollution in scenarios with infrequent data collection such as weekly or monthly. Here we showed sparse-dataset-based prediction of water pollution using machine learning. We investigated the efficacy of a traditional Recurrent Neural Network alongside three Long Short-Term Memory (LSTM) models, integrated with the Load Estimator (LOADEST). The research was conducted at a river-lake confluence, an area with intricate hydrological patterns. We found that the Self-Attentive LSTM (SA-LSTM) model outperformed the other three machine learning models in predicting water quality, achieving Nash-Sutcliffe Efficiency (NSE) scores of 0.71 for CODMn and 0.57 for NH3N when utilizing LOADEST-augmented water quality data (referred to as the SA-LSTM-LOADEST model). The SA-LSTM-LOADEST model improved upon the standalone SA-LSTM model by reducing the Root Mean Square Error (RMSE) by 24.6% for CODMn and 21.3% for NH3N. Furthermore, the model maintained its predictive accuracy when data collection intervals were extended from weekly to monthly. Additionally, the SA-LSTM-LOADEST model demonstrated the capability to forecast pollution loads up to ten days in advance. This study shows promise for improving water quality modeling in regions with limited monitoring capabilities.

Study region: Chao Phraya River Basin—a major river with unique characteristics located in Thailand. Study focus: This study sought to simulate the flow rates in the Chao Phraya River Basin, which is a tidal river that poses challenges to traditional modeling approaches. The soil and water assessment tool (SWAT) is a hydrological model extensively employed for simulating flow rates. However, limitations arise in applying the SWAT model to the Chao Phraya River Basin due to its tidal nature, resulting in an unsatisfactory model performance. To address this, a long short-term memory (LSTM) model, i.e., the SWAT–LSTM model, was introduced to complement the SWAT model. New hydrological insights for the Region: The collaborative coupling of hydrological information derived from the SWAT and LSTM notably enhanced the model performance. This improvement was assessed using the Nash-Sutcliffe efficiency (NSE), demonstrating an increase from 0.13 to 0.72. The incorporation of topographic static data in the coupling model was also investigated to provide the basic characteristics of the basin to the model. The results yielded an NSE exceeding 0.79. The shoreline water level was identified as a crucial input feature for indicating tidal patterns. The findings highlight the effectiveness of coupling the SWAT with LSTM for predicting tidal river flow rates, implying their applicability in similar scenarios across different basins.

Study region: Eight governorates in upper Egypt namely Aswan, Asyut, Beni-Suef, Fayoum, Luxor, Minya, Qena and Sohag. Study focus: This study aims to develop novel hybrid machine learning (ML) models for forecasting the drought phenomena based on limited inputs for the eight Egyptian govern-orates, and ii) evaluate the performance and accuracy of the developed ML models for predicting Palmer Drought Severity Index (PDSI) to recommend the optimal model based on performance statistical metrics. The hybrid ML models were Convolution Neural Networks (CNN)-Long Short-Term Memory (LSTM), CNN-Random Forest (RF), CNN-Support Vector Machine (SVR), and CNN-Extreme Gradient Boosting (XGB). New hydrological insights for the region: Results showed that CNN-LSTM model outperformed the others followed by CNN-RF. Values of NSE, MAE, MARE, IA, R2, and RMSE for CNN-LSTM were 0.885, 0.915, − 2.073, 0.967, 0.885, and 0.573, respectively. For the testing stage CNN-SVR model was found to perform the best; average values of NSE, MAE, MARE, IA, R2, and RMSE were 0.828, 0.364, − 2.903, 0.950, 0.828 and 0.688, respectively. This study provided a way forward for convenient estimation of the PDSI Index from the meteorological data in terms of advancing deep learning algorithms. The developed hybrid models, more or less, can satisfactory predict PDSI values. Additionally, the study suggests the CNN-LSTM model as the most suitable model to advance future investigation in the study area.

The world's need for energy is rising due to factors like population growth, economic expansion, and technological breakthroughs. However, there are major consequences when gas and coal are burnt to meet this surge in energy needs. Although these fossil fuels are still essential for meeting energy demands, their combustion releases a large amount of carbon dioxide and other pollutants into the atmosphere. This significantly jeopardizes community health in addition to exacerbating climate change, thus it is essential need to move swiftly to incorporate renewable energy sources by employing advanced information and communication technologies. However, this change brings up several security issues emphasizing the need for innovative cyber threats detection and prevention solutions. Consequently, this study presents bigdata sets obtained from the solar and wind powered distributed energy systems through the blockchain-based energy networks in the smart grid (SG). A hybrid machine learning (HML) model that combines both the Deep Learning (DL) and Long-Short-Term-Memory (LSTM) models characteristics is developed and applied to identify the unique patterns of Denial of Service (DoS) and Distributed Denial of Service (DDoS) cyberattacks in the power generation, transmission, and distribution processes. The presented big datasets are essential and significantly helps in identifying and classifying cyberattacks, leading to predicting the accurate energy systems behavior in the SG.

Parkinson's disease (PD) is a widespread neurodegenerative condition that affects many individuals annually. Early identification and monitoring of disease progression are crucial to effectively managing symptoms and preventing motor complications. This research proposes an automated PD diagnosis and severity-grading model based on time-frequency and fuzzy features using improved uni-directional and bi-directional long short-term memory networks with sensitive hyperparameters optimization. We utilize vertical ground reaction force signals collected from Physionet's publicly available dataset recorded during regular and dual-task clinical trials of walking measurements. Only the cumulative signal of both feet was then utilized and segmented into 30-s windows without further pre-processing. Subsequently, we extracted only four key time-frequency and fuzzy features from each segment, effectively capturing the signal's inherent uncertainty. Bayesian optimization is employed in both detection and grading approaches to fine-tune the two critical hyperparameters: the initial learning rate and the number of hidden units in the network. The detection phase yields an exceptional accuracy of 99.19%, surpassing state-of-the-art studies with the same dataset. In the grading phase, classification based on the unified PD rating scale values achieves an accuracy of 92.28%. The proposed study delves into the potential of cumulative gait signals as a powerful diagnostic tool for PD, aiming to extract precise and intricate information by implementing straightforward and minimal processing endeavors. This method demonstrates significant efficiency in terms of complexity, cost, and energy consumption by utilizing a single-dimensional signal, eliminating the need for pre-processing steps, and limiting the features used for training.

Hand Gesture Recognition (HGR) using electromyography (EMG) signals is a challenging problem due to the variability and noise in the signals across individuals. This study addresses this challenge by examining the effect of incorporating a post-processing algorithm, which filters the sequence of predictions and removes spurious labels, on the performance of a HGR model based on spectrograms and Convolutional Neural Networks (CNN). The study also compares CNN vs CNN-LSTM to assess the influence of the memory cells on the model. The EMG-EPN-612 dataset, which contains measurements of EMG signals for 5 hand gestures from 612 subjects, was used for training and testing. The results showed that the post-processing algorithm increased the recognition accuracy by 41.86% for the CNN model and 24.77% for the CNN-LSTM model. The inclusion of the memory cells increased accuracy by 3.29%, but at the cost of 53 times more learnables. The CNN-LSTM model with post-processing achieved a mean recognition accuracy of 90.55% (SD=9.45%). These findings suggest new paths for research in HGR architectures beyond the traditional focus on the classification and feature extraction stages. For reproducibility purposes, we made publicly available the source code in Github.

The rapid rate of transformation in the power sector of India has placed a significant emphasis on robust grids and distributed generation units. The observable shift in the energy sector, especially in wind and solar energy, also requires smooth integration of Distributed Generation units with the existing power grid. Precise wind power generation forecast, therefore, becomes an important and complex task for the strategic planning and management of the systems. We, thus, aim towards a system that can actually provide precise wind power forecasts by applying machine learning techniques. This work proposes a comparative and comprehensive analysis of Artificial Neural Network (ANN), Recurrent Neural Network (RNN), Long Short-Term Memory (LSTM), Gated Recurrent Unit (GRU), and Autoregressive Integrated Moving Average (ARIMA) model. The experimentations and modelling are performed considering meteorological and historical power generation data. The study is concentrated in Kutch, Gujarat and is validated on the data collected from the Central Electricity Authority (CEA), India for power generation data and weather data collected from regional weather centres. The findings show that ARIMA outperforms the other models for non-linear data in multivariate analysis, with a MAPE score of less than 5.5 on the prediction dataset.

Long Short-Term Memory (LSTM) is a popular Recurrent Neural Network (RNN) algorithm known for its ability to effectively analyze and process sequential data with long-term dependencies. Despite its popularity, the challenge of effectively initializing and optimizing RNN-LSTM models persists, often hindering their performance and accuracy. This study presents a systematic literature review (SLR) using an in-depth four-step approach based on the PRISMA methodology, incorporating peer-reviewed articles spanning 2018–2023. It aims to address how weight initialization and optimization techniques can bolster RNN-LSTM performance. This SLR offers a detailed overview across various applications and domains, and stands out by comprehensively analyzing modeling techniques, datasets, evaluation metrics, and programming languages associated with these networks. The findings of this SLR provide a roadmap for researchers and practitioners to enhance RNN-LSTM networks and achieve superior results.

Selecting appropriate input variables for developing a rainfall prediction model is significantly difficult. The present study proposed an innovative framework for input variable selection (IVS) model bootstrapped long short-term recurrent neural network (BTSP-LSTM-RNN) to identify relevant variables for monthly rainfall forecasting. Monthly meteorological and large-scale climatic variables (LCVs) from 1993 to 2022 at two selected river basins in the northern region of Thailand were used for model development. The proposed BTSP-LSTM-RNN model results were compared with the support vector regression with recursive feature elimination (SVR-RFE) and Gradient boosting (GB) by statistical metrics such as coefficient of determination (R2), mean absolute error (MAE), relative root mean squared error (RRMSE), Pearson’s correlation coefficient (r) and mean absolute percentage error (MAPE). BTSP-LSTM-RNN demonstrated exceptional performance, boasting a higher R2 (0.84), MAE (92.28), RRMSE (10.36) in the Wang basin, and R2 (0.83), MAE (242.60), RRMSE (9.93) in the Nan basin. BTSP-LSTM-RNN also achieved the lowest MAPE of 29.82%. Based on this IVS model results, two input variable combinations (IVCs) were designed. IVC-1 is based on BTSP-LSTM-RNN selection, and IVC-2 is an original set of variables. LSTM-RNN, multi-layer perceptron artificial neural network (MLP-ANN), and ensemble model with bootstrapping on random forest (RF) were employed for monthly prediction. When BTSP-LSTM-RNN's selected input variables from IVC-1 are utilized, the BSTP-RF model demonstrates robust performance. It achieves a high R2 (0.82), a low RRMSE (10.13%) suggests accurate predictions, and the r of 0.91 further supports the model's strong linear relationship with observed rainfall data. Based on prediction model results, the BTSP-LSTM-RNN (IVS) model plays a pivotal role in the selection of input variables for rainfall forecasting and its impact on the performance of prediction models (BSTP-RF, MLP-ANN, and LSTM-RNN). These results consistently underscored the pivotal role of the BTSP-LSTM-RNN IVS model in enhancing the precision and reliability of rainfall predictions.

With the construction of new-type power system under the ”double carbon” target and the increasing diversification of the energy demand of the user side, the short-term load forecasting of power system is facing new challenges. In order to fully exploit the massive information contained in big data, this paper proposed a new short-term load forecasting method for new-type power system considering multiple factors, which based on Graph Convolutional Network (GCN) and Long Short-Term Memory network (LSTM). Spearman rank correlation coefficient was used to analyze the correlation between load and meteorological factors, and a quantitative model including meteorological factors, date factors and regional factors was established. Thus, GCN and LSTM were jointly used to extract the spatial and temporal characteristics of massive data respectively, and finally the short-term power load forecasting was achieved. The public data sets were used for performance verification compared with three comparison models, LSTM, CNN-LSTM and TCN-LSTM. The results show that the proposed method can make full use of the influence of multi-dimensional data, meanwhile improve the load prediction accuracy and training efficiency effectively.

Transmission Expansion Planning (TEP) can be essentially affected by solar Photovoltaic (PV) penetration because its generation is uncertain depending on weather and the environment at a specific time. This uncertainty can impact transmission expansion planning because the uncertain generation of solar PV causes difficulty to expand transmission lines with secure system operations. The aim of the proposed TEP method is to construct transmission lines to handle the increase in solar PV sources and electricity demand while ensuring that the power system constraints are not violated. The goal of the objective function is to minimize the total cost of the system, which comprises both the investment cost and the operating cost of generation. Hourly resolution values of solar PV generation and electricity demand are considered in the planning. Moreover, to obtain accurate values of solar PV generation for planning a system with secure operations, this paper proposes a Long Short-Term Memory (LSTM) for forecasting solar PV generation. The Binary Differential Evolution (BDE) algorithm is applied to solve the TEP problem. The proposed TEP method is tested through an IEEE 24 bus test system. The results from the proposed method show that the obtained expansion plan provides the minimum total system cost without system violations.

In this work, a multivariate time-series machine learning meta-model is developed to predict the transient response of a typical nuclear power plant (NPP) undergoing a steam generator tube rupture (SGTR). The model employs Recurrent Neural Networks (RNNs), including the Long Short-Term Memory (LSTM), Gated Recurrent Unit (GRU), and a hybrid CNN-LSTM model. To address the uncertainty inherent in such predictions, a Bayesian Neural Network (BNN) was implemented. The models were trained using a database generated by the Best Estimate Plus Uncertainty (BEPU) methodology; coupling the thermal hydraulics code, RELAP5/SCDAP/MOD3.4 to the statistical tool, DAKOTA, to predict the variation in system response under various operational and phenomenological uncertainties. The RNN models successfully captures the underlying characteristics of the data with reasonable accuracy, and the BNN-LSTM approach offers an additional layer of insight into the level of uncertainty associated with the predictions. The results demonstrate that LSTM outperforms GRU, while the hybrid CNN-LSTM model is computationally the most efficient. This study aims to gain a better understanding of the capabilities and limitations of machine learning models in the context of nuclear safety. By expanding the application of ML models to more severe accident scenarios, where operators are under extreme stress and prone to errors, ML models can provide valuable support and act as expert systems to assist in decision-making while minimizing the chances of human error.

Predicting future fouling status is a crucial but tough topic in the applications of energy conservation and pollution reduction at coal-fired power plants because of the significant influence that ash slag has on the heat transfer efficiency of boilers in coal-fired power plants. For the prediction of gray areas in heated areas, a hybrid system based on complementary ensemble empirical modal decomposition, gray models, and long short-term memory networks is presented. This is because the time series of ash pollution degrees is not linear and not smooth. Initially, using a complementary ensemble empirical modal decomposition, the original sequence after wavelet threshold denoising is divided into a number of subseries components. The projected values for the cleanliness factor were then generated by superimposing the predictions from the IMF and residual components. The experimental findings support the model’s precision and dependability and demonstrate that the CEEMD-GM-LSTM model does, in fact, perform very well in forecasting the ash situation in the heated zone.

The smart grid is an innovation that employs two-way communications to give innovative services to end consumers. Due to the severe contradictions in this connection, this system may be the target of numerous cyber-attacks. Intelligent grid networks can be protected by employing intrusion detection systems. IDS increases smart grid security by identifying malicious activity in the networks. However, current detection methods have several areas for improvement, including a high false alarm rate and low detection accuracy. The paper proposes an innovative intrusion detection strategy for intelligent grids combining DL-based and feature-based techniques. For this, the dataset is pre-processed, and pre-processing is done by utilizing min–max normalization. Then, features including mean, median, mode, standard deviation, information gain, mutual information, correlation coefficient, data percentiles, and autoregressive data are extracted. Next, African Vulture Optimization Algorithm organizes feature selection. Finally, DBN-LSTM is utilized for categorization to identify normal and attack packets. The developed method attains higher performance when compared with other existing techniques. Hence, the outcomes demonstrate that the AVOA-DBN-LSTM technique has a reliable potential for cybersecurity intrusion detection.

The risk warning for steady-state power quality in the power grid is essential for its prevention and management. However, current risk warning methods fall short in predicting the power quality trend while accounting for potential risks. Consequently, this study introduces a novel steady-state power quality risk warning method utilizing VMD-LSTM and a fuzzy model. Firstly, a power quality index prediction method based on variational mode decomposition (VMD) and long short-term memory (LSTM) is proposed. This approach significantly enhances prediction accuracy. Secondly, a power quality risk warning method incorporating kernel density estimation (KDE) and a fuzzy model is proposed, which systematically addresses the uncertainty associated with power quality risks. To validate the effectiveness and practicality of the proposed method, experiments are conducted using field monitoring data from a residential load in southern China. The results affirm the reliability and applicability of the proposed method. The simulation results show that the median error of prediction of power quality indexes by the proposed method is 5.03 % during the evaluated time period, and the prediction accuracy is mostly maintained above 90 %.

Study region: Four sub-basins of the Songhua River basin, northeast China. Study focus: Conventional runoff models typically require in-depth knowledge of the hydrological and physical processes and are costly to construct and compute. Moreover, these models predominantly rely on ground site data, where incomplete or delayed data might introduce simulation uncertainty. Therefore, it is imperative to provide a scientifically rigorous and rational approach for simulating the runoff process, effectively addressing the limitations of existing methods. Combining a long short-term memory (LSTM) network with a modified Michel soil conservation service (MMSCS) algorithm, this study proposed the LSTM-MMSCS runoff simulation scheme. New hydrological insights for the region: The LSTM-MMSCS model was constructed by adjusting and optimizing the difference characteristics of the LSTM runoff simulation by establishing regression relationships according to the MMSCS-calculated runoff depth. LSTM-MMSCS adopted the coupling method of hydrological mechanism and deep learning to establish a simulation framework with adaptive feedback and adjustment between observed and simulated data. This scheme incorporated satellite meteorological products, solving the problem of inaccuracies caused by standard models' ineffective mining of temporal series information. LSTM-MMSCS reduced overall runoff error (RMSE was reduced from 50.07 mm to 24.47 mm) and effectively alleviated the problem of peak runoff underestimation (the relative error was reduced from 30.39% to 13.39%) compared to LSTM. Using satellite meteorological data to drive LSTM-MMSCS enabled runoff change trends visualization and aids in abnormal runoff localization.

This article aims to propose short-term forecasting for a combined model of the hourly electricity price and the hourly load. Where both the independent system operators and the market participants must deal with the uncertainty and volatility of the day-ahead electricity price. Therefore, an accurate prediction model has been needed for a combined short-term forecasting model for both the hourly electricity price and the hourly load in the deregulated market. To get a better prediction, real-time data for the day-ahead electricity price and the hourly load are provided from the ISO New England Control Area (ISO-NE-CA) market. This study presents the combined model based on four featured machine and deep learning algorithms: Feed Forward Artificial Neural Network, Adaptive Neuro-Fuzzy Inference System, Long Short-Term Memory, and Gated Recurrent Units. The results are obtained from two scenarios, In the first scenario, the data has been trained and tested without depending on other factors such as temperatures and calendar. In the second scenario, the data has been trained and tested depending on the previously mentioned factors. The obtained results from the four algorithms have been compared to show their performance and their efficiency in the values of Mean Absolute Error (MAE), Root Mean Squared Error (RMSE), Normalized Root Mean Squared Error (NRMSE), and Mean Absolute Percentage Error (MAPE). The four featured algorithms prove their good performance with less error in the forecasting model.

Air pollution threatens human health, necessitating effective and convenient air quality monitoring. Recently, there has been a growing interest in using camera images for air quality estimation. However, a major challenge has been nighttime detection due to the limited visibility of nighttime images. Here we present a hybrid deep learning model, capitalizing on the temporal continuity of air quality changes for estimating outdoor air quality from surveillance images. Our model, which integrates a convolutional neural network (CNN) and long short-term memory (LSTM), adeptly captures spatial-temporal image features, enabling air quality estimation at any time of day, including PM2.5 and PM10 concentrations, as well as the air quality index (AQI). Compared to independent CNN networks that solely extract spatial features, our model demonstrates superior accuracy on self-constructed datasets with R2 = 0.94 and RMSE = 5.11 μg m−3 for PM2.5, R2 = 0.92 and RMSE = 7.30 μg m−3 for PM10, and R2 = 0.94 and RMSE = 5.38 for AQI. Furthermore, our model excels in daytime air quality estimation and enhances nighttime predictions, elevating overall accuracy. Validation across diverse image datasets and comparative analyses underscore the applicability and superiority of our model, reaffirming its applicability and superiority for air quality monitoring.