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

Machine learning algorithms have proven to be practical in a wide range of applications. Many studies have been conducted on the operational energy consumption and thermal comfort of radiant floor systems. This paper conducts a case study in a self-designed experimental setup that combines radiant floor and fan coil cooling (RFCFC) and develops a data monitoring system as a source of historical operational data. Seven machine learning algorithms (extreme learning machine (ELM), convolutional neural network (CNN), genetic algorithm-back propagation (GA-BP), radial basis function (RBF), random forest (RF), support vector machine (SVM), and long short-term memory (LSTM)) were employed to predict the behavior of the RFCFC system. Corresponding prediction models were then developed to evaluate operative temperature (T_op) and energy consumption (E_h). The performance of the model was evaluated using five error metrics. The obtained results showed that the RF model had very high performance in predicting T_op and E_h, with high correlation coefficients (>0.9915) and low error metrics. Compared with other models, it also demonstrated high accuracy in E_h prediction, yielding maximum reductions of 68.1, 82.4, and 43.2% in the mean absolute percentage error (MAPE), mean squared error (MSE), and mean absolute error (MAE), respectively. A sensitivity ranking algorithm analysis was also conducted. The obtained results demonstrated the importance of adjusting parameters, such as the radiant floor supply water temperature, to enhance the indoor comfort. This study provides a novel and effective method for evaluating the energy efficiency and thermal comfort of radiant cooling systems. It also provides insights for optimizing the efficiency and thermal comfort of RFCFC systems, and lays a theoretical foundation for future studies integrating machine learning algorithms in this field.

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

Accurate cooling load prediction is essential for energy-efficient HVAC system operation. However, the stochastic and nonlinear nature of load data challenges conventional neural networks, causing prediction delays and errors. To address this, a novel hybrid model is developed. The approach first applies a two-stage decomposition (CEEMDAN with K-means and VMD) to process complex cooling load data. Then, a CNN-BiLSTM network optimized by the Crested Porcupine Optimizer and integrated with an attention mechanism is constructed for prediction. Experimental results demonstrate the model’s high performance, achieving a 96.75% prediction accuracy with a MAPE of 3.25% and an R² of 0.9929. The proposed model shows strong robustness and generalization, providing a reliable reference for intelligent building energy management.