An air-conditioning system (ACS), which consumes large amounts of high-grade energy, is essential for maintaining the indoor thermal environment of modern buildings. However, an ACS consumes almost half of the total energy of the building. Therefore, it is necessary to reduce the energy consumption of the ACS to promote energy conservation and emission reduction in the building sector. In fact, there is an abundance of waste heat and low-grade energies with the potential to be utilized in ACS in nature, but many of them are not utilized efficiently or cannot be utilized at all due to the low efficiency of thermal energy conversion. Known as a passive thermal transfer device, the application of a heat pipe (HP) in the ACS has shown explosive growth in recent years. HPs have been demonstrated to be an effective method for reducing building cooling and heating demands and energy consumption in ACS with experimental and simulation methods. This paper summarizes the different HP types applied in the ACS and provides brief insight into the performance enhancement of the ACS integrated with HP. Four types of HPs, namely tubular HP (THP), loop HP (LHP), pulsating HP (PHP) and flat HP (FHP), are presented. Their working principles and scope of applications are reviewed. Then, HPs used in natural cooling system, split air conditioner (SAC), centralized ACS (CACS) and cooling terminal devices are comprehensively reviewed. Finally, the heat transfer characteristics and energy savings of the above systems are critically analyzed. The results show that the performance of the HP is greatly affected by its own structure, working fluid and external environmental conditions. The energy saving of ACS coupled with HP is 3–40.9%. The payback period of this system ranges from 1.9–10 years. It demonstrates that the HP plays a significant role in reducing ACS energy consumption and improving indoor thermal comfort.

Artificial intelligence algorithms have gained widespread adoption in the field of air conditioning load prediction. However, their prediction accuracy is substantially influenced by the quality of training samples. Training samples that lack relevance to the predicted moments can introduce interference into the neural network's learning process, potentially leading to a state of local convergence during its iterative process. This, in turn, diminishes the robustness and generalization capabilities of the prediction model. To enhance the prediction accuracy of air conditioning load prediction models based on artificial intelligence algorithms, this study presents an artificial intelligence algorithm prediction model based on the method of sample similarity sample screening. Initially, the comprehensive similarity coefficient between samples is computed using the gray correlation analysis method, enriched with enhancements in information entropy. Subsequently, a subset of closely related samples is extracted from the original dataset and employed as the training dataset for the artificial intelligence prediction model. Finally, the trained artificial intelligence algorithm prediction model is deployed for air conditioning load prediction. The results illustrate that the method of screening training samples based on sample similarity effectively improves the prediction accuracy of BP neural network (BPNN) and extreme learning machine (ELM) prediction models. However, it is important to note that this approach may not be suitable for genetic algorithm BPNN (GABPNN) and support vector regression (SVR) models.