Enhancing Air Conditioning System Efficiency Through Load Prediction and Deep Reinforcement Learning: A Case Study of Ground Source Heat Pumps

Energies, Год журнала: 2025, Номер 18(1), С. 199 - 199

Опубликована: Янв. 5, 2025

This study proposes a control method that integrates deep reinforcement learning with load forecasting, to enhance the energy efficiency of ground source heat pump systems. Eight machine models are first developed predict future cooling loads, and optimal one is then incorporated into learning. Through interaction environment, strategy identified using Q-network optimize supply water temperature from source, allowing for savings. The obtained results show XGBoost model significantly outperforms other in terms prediction accuracy, reaching coefficient determination 0.982, mean absolute percentage error 6.621%, variation root square 10.612%. Moreover, savings achieved through forecasting-based greater than those traditional constant methods by 10%. Additionally, without shortening interval, improved 0.38% compared do not use predictive information. approach requires only continuous between agent which makes it an effective alternative scenarios where sensor equipment data present. It provides smart adaptive optimization solution heating, ventilation, air conditioning systems buildings.

Язык: Английский

A Systematic Review of Building Energy Consumption Prediction: From Perspectives of Load Classification, Data-Driven Frameworks, and Future Directions

Applied Sciences, Год журнала: 2025, Номер 15(6), С. 3086 - 3086

Опубликована: Март 12, 2025

The rapid development of machine learning and artificial intelligence technologies has promoted the widespread application data-driven algorithms in field building energy consumption prediction. This study comprehensively explores diversified prediction strategies for different time scales, types, forms, constructing a framework this field. With process as core, it deeply analyzes four key aspects data acquisition, feature selection, model construction, evaluation. review covers three acquisition methods, considers seven factors affecting loads, introduces efficient extraction techniques. Meanwhile, conducts an in-depth analysis mainstream models, clarifying their unique advantages applicable scenarios when dealing with complex data. By systematically combing existing research, paper evaluates advantages, disadvantages, applicability each method provides insights into future trends, offering clear research directions guidance researchers.

Язык: Английский

Advances in fault detection techniques for automated manufacturing systems in industry 4.0

Frontiers in Mechanical Engineering, Год журнала: 2025, Номер 11

Опубликована: Апрель 17, 2025

Fault detection and diagnosis are essential for maintaining the continuous operation of manufacturing systems. To achieve this, an innovative tool is required to immediately identify any faults in production process recommend appropriate mechanisms be adopted proactively prevent future mishaps or accidents. This capability critical many industries improve efficiency effectiveness their processes. Several methods can used detect trends patterns given determine if variable within normal limits. However, these techniques may only evident characteristics defects while leaving behind latent ones. paper aims review recent achievements classics fault detection, suggest steps that taken plan implement this process. It will also explore emerging research streams, issues field, strategies applied overcome barriers. The outlines how performance diagnostics improved processes a safer fully efficient environment promoted.

Язык: Английский