Optimizing load demand forecasting in educational buildings using quantum-inspired particle swarm optimization (QPSO) with recurrent neural networks (RNNs):a seasonal approach DOI Creative Commons

Sunawar Khan,

Tehseen Mazhar, Tariq Shahzad

и другие.

Scientific Reports, Год журнала: 2025, Номер 15(1)

Опубликована: Июнь 3, 2025

This study uses Quantum Particle Swarm Optimization (QPSO) optimized Recurrent Neural Networks (RNN), standard RNN, and autoregressive integrated moving average (ARIMA) models to anticipate educational building power demand accurately. Energy efficiency, cost reduction, resource allocation depend on accurate load forecasts. The evaluates model performance using year-long data from seasonal, daily, hourly fluctuations. Performance indicators, including Mean Absolute Error (MAE), Squared (MSE), Root (RMSE), were used assess the models. QPSO-optimized RNN outperformed traditional ARIMA with lowest MAE of 15.2, MSE 520.15, RMSE 22.8. Comparative investigation shows QPSO-RNN's capacity capture complicated patterns, especially during peak demand. that hybrid optimization can improve forecasting accuracy, making it a powerful tool for energy management in dynamic contexts.

Язык: Английский

Exploring commuter stress dynamics through machine learning and double optimization DOI Creative Commons
Ashar Ahmed, Mario Muñoz-Organero, Bushra Aijaz

и другие.

Mehran University Research Journal of Engineering and Technology, Год журнала: 2025, Номер 44(2), С. 35 - 46

Опубликована: Апрель 9, 2025

Travel dynamics significantly impact commuter stress, influenced by traffic behavior, road conditions, travel modes, distance, and socio-demographic characteristics. Previous research on stress often exhibits limitations, including narrow scopes focusing specific routes, vehicle types, or demographics. This study addresses these constraints employing a comprehensive approach to analyze the influence of various attributes levels. An interview-based dataset was collected capture multifaceted experiences users. Five tree-based machine learning models–Decision Tree (DT), Random Forests (RF), Extra Trees (ET), Extreme Gradient Boosting (XGBoost), k-Nearest Neighbor (k-NN)–were deployed for imbalanced multi-class classification. XGBoost demonstrated superior performance with highest accuracy (73.33%) precision (75.63%) standard deviation ±5.9. A novel double hyperparameter optimization technique enhanced prediction across all models, notably increasing k-NN classifier’s 19.99%. The SHAP (SHapley Additive exPlanations) method utilized model interpretability, revealing distance traveled per day as most influential factor levels, followed mode transport, gender, age low, medium, high-stress categories, respectively. also examines features individual levels through random instance selection. provides valuable insights into complex interplay between paving way development effective mitigation strategies improved

Язык: Английский

Процитировано

0
Optimizing load demand forecasting in educational buildings using quantum-inspired particle swarm optimization (QPSO) with recurrent neural networks (RNNs):a seasonal approach DOI Creative Commons

Sunawar Khan,

Tehseen Mazhar, Tariq Shahzad

и другие.

Scientific Reports, Год журнала: 2025, Номер 15(1)

Опубликована: Июнь 3, 2025

This study uses Quantum Particle Swarm Optimization (QPSO) optimized Recurrent Neural Networks (RNN), standard RNN, and autoregressive integrated moving average (ARIMA) models to anticipate educational building power demand accurately. Energy efficiency, cost reduction, resource allocation depend on accurate load forecasts. The evaluates model performance using year-long data from seasonal, daily, hourly fluctuations. Performance indicators, including Mean Absolute Error (MAE), Squared (MSE), Root (RMSE), were used assess the models. QPSO-optimized RNN outperformed traditional ARIMA with lowest MAE of 15.2, MSE 520.15, RMSE 22.8. Comparative investigation shows QPSO-RNN's capacity capture complicated patterns, especially during peak demand. that hybrid optimization can improve forecasting accuracy, making it a powerful tool for energy management in dynamic contexts.

Язык: Английский

Процитировано

0