Decomposition aided attention-based recurrent neural networks for multistep ahead time-series forecasting of renewable power generation

PeerJ Computer Science, Год журнала: 2024, Номер 10, С. e1795 - e1795

Опубликована: Янв. 18, 2024

Renewable energy plays an increasingly important role in our future. As fossil fuels become more difficult to extract and effectively process, renewables offer a solution the ever-increasing demands of world. However, shift toward renewable is not without challenges. While reliable means storage that can be converted into usable energy, are dependent on external factors used for generation. Efficient often relying batteries have limited number charge cycles. A robust efficient system forecasting power generation from sources help alleviate some difficulties associated with transition energy. Therefore, this study proposes attention-based recurrent neural network approach generated sources. To networks make accurate forecasts, decomposition techniques utilized applied time series, modified metaheuristic introduced optimized hyperparameter values networks. This has been tested two real-world datasets covering both solar wind farms. The models by metaheuristics were compared those produced other state-of-the-art optimizers terms standard regression metrics statistical analysis. Finally, best-performing model was interpreted using SHapley Additive exPlanations.

Язык: Английский

---

The explainable potential of coupling hybridized metaheuristics, XGBoost, and SHAP in revealing toluene behavior in the atmosphere

The Science of The Total Environment, Год журнала: 2024, Номер 929, С. 172195 - 172195

Опубликована: Апрель 15, 2024

Язык: Английский

---

Marine Vessel Classification and Multivariate Trajectories Forecasting Using Metaheuristics-Optimized eXtreme Gradient Boosting and Recurrent Neural Networks

Applied Sciences, Год журнала: 2023, Номер 13(16), С. 9181 - 9181

Опубликована: Авг. 11, 2023

Maritime vessels provide a wealth of data concerning location, trajectories, and speed. However, while these are meticulously monitored logged to maintain course, they can also meta information. This work explored the potential data-driven techniques applied artificial intelligence (AI) tackle two challenges. First, vessel classification was through use extreme gradient boosting (XGboost). Second, trajectory time series forecasting tackled long-short-term memory (LSTM) networks. Finally, due strong dependence AI model performance on proper hyperparameter selection, boosted version well-known particle swarm optimization (PSO) algorithm introduced specifically for tuning hyperparameters models used in this study. The methodology real-world automatic identification system (AIS) both marine forecasting. Boosted PSO (BPSO) compared contemporary optimizers showed promising outcomes. XGBoost tuned using attained an overall accuracy 99.72% problem, LSTM mean square error (MSE) 0.000098 prediction challenge. A rigid statistical analysis performed validate outcomes, explainable principles were determined best-performing models, gain better understanding feature impacts decisions.

Язык: Английский

---

Potential of Coupling Metaheuristics-Optimized-XGBoost and SHAP in Revealing PAHs Environmental Fate

Toxics, Год журнала: 2023, Номер 11(4), С. 394 - 394

Опубликована: Апрель 21, 2023

Polycyclic aromatic hydrocarbons (PAHs) refer to a group of several hundred compounds, among which 16 are identified as priority pollutants, due their adverse health effects, frequency occurrence, and potential for human exposure. This study is focused on benzo(a)pyrene, being considered an indicator exposure PAH carcinogenic mixture. For this purpose, we have applied the XGBoost model two-year database pollutant concentrations meteorological parameters, with aim identify factors were mostly associated observed benzo(a)pyrene describe types environments that supported interactions between other polluting species. The data collected at energy industry center in Serbia, vicinity coal mining areas power stations, where maximum concentration period reached 43.7 ngm-3. metaheuristics algorithm has been used optimize hyperparameters, results compared models tuned by eight cutting-edge algorithms. best-produced was later interpreted applying Shapley Additive exPlanations (SHAP). As indicated mean absolute SHAP values, temperature surface, arsenic, PM10, total nitrogen oxide (NOx) appear be major affecting its environmental fate.

Язык: Английский

---

Explainable Artificial Intelligence Visions on Incident Duration Using eXtreme Gradient Boosting and SHapley Additive exPlanations

Multimodal Transportation, Год журнала: 2025, Номер unknown, С. 100209 - 100209

Опубликована: Фев. 1, 2025

Язык: Английский

---

Computer-Vision Unmanned Aerial Vehicle Detection System Using YOLOv8 Architectures

International Journal of Robotics and Automation Technology, Год журнала: 2024, Номер 11, С. 1 - 12

Опубликована: Май 22, 2024

Abstract: This work aims to test the performance of you only look once version 8 (YOLOv8) model for problem drone detection. Drones are very slightly regulated and standards need be established. With a robust system detecting drones possibilities regulating their usage becoming realistic. Five different sizes were tested determine best architecture size this problem. The results indicate high across all models that each is used specific case. Smaller suited lightweight approaches where some false identification tolerable, while largest with stationary systems require precision.

Язык: Английский

---

Improved intelligent methods for power transformer fault diagnosis based on tree ensemble learning and multiple feature vector analysis

Electrical Engineering, Год журнала: 2023, Номер 106(3), С. 2575 - 2594

Опубликована: Ноя. 3, 2023

Язык: Английский

---

XGBoost Tuned by Hybridized SCA Metaheuristics for Intrusion Detection in Healthcare 4.0 IoT Systems

Algorithms for intelligent systems, Год журнала: 2024, Номер unknown, С. 1 - 16

Опубликована: Янв. 1, 2024

Язык: Английский

---

Short-term forecasting of vegetable prices based on LSTM model—Evidence from Beijing’s vegetable data

PLoS ONE, Год журнала: 2024, Номер 19(7), С. e0304881 - e0304881

Опубликована: Июль 11, 2024

The vegetable sector is a vital pillar of society and an indispensable part the national economic structure. As significant segment agricultural market, accurately forecasting prices holds importance. Vegetable market pricing subject to myriad complex influences, resulting in nonlinear patterns that conventional time series methodologies often struggle decode. In this paper, we exploit average daily price data six distinct types vegetables sourced from seven key wholesale markets Beijing, spanning 2009 2023. Upon training LSTM model, discovered it exhibited exceptional performance on test dataset. Demonstrating robust predictive across various categories, model shows commendable generalization abilities. Moreover, has higher accuracy compared several machine learning methods, including CNN-based approaches. With R2 score 0.958 MAE 0.143, our registers enhancement over 5% forecast relative counterparts. Therefore, by predicting for upcoming week, envision application real-world settings aid growers, consumers, policymakers facilitating informed decision-making. insights derived research could augment transparency optimize supply chain management. Furthermore, contributes stability balance demand, offering valuable reference sustainable development industry.

Язык: Английский

---

Yield prediction for crops by gradient-based algorithms

PLoS ONE, Год журнала: 2024, Номер 19(8), С. e0291928 - e0291928

Опубликована: Авг. 26, 2024

A timely and consistent assessment of crop yield will assist the farmers in improving their income, minimizing losses, deriving strategic plans agricultural commodities to adopt import-export policies. Crop predictions are one various challenges faced agriculture sector play a significant role planning decision-making. Machine learning algorithms provided enough belief proved ability predict yield. The selection most suitable is influenced by environmental factors such as temperature, soil fertility, water availability, quality, seasonal variations, well economic considerations stock preservation capabilities, market demand, purchasing power, prices. paper outlines framework used evaluate performance machine-learning for forecasting yields. models were based on range prime parameters including pesticides, rainfall average temperature. Results three machine algorithms, Categorical Boosting (CatBoost), Light Gradient-Boosting (LightGBM), eXtreme Gradient (XGBoost) compared found more accurate than other predicting RMSE R 2 values calculated compare predicted observed rice yields, resulting following values: CatBoost with 800 (0.24), LightGBM 737 (0.33), XGBoost 744 (0.31). Among these demonstrated highest precision achieving an accuracy rate 99.123%.

Язык: Английский

---