Abstract

Data-driven fault inference in building HVAC systems faces a critical challenge due to the absence of labeled operational data in the target working condition. To address this, transfer learning-based methods are proposed to leverage inference knowledge from labeled datasets acquired in distinct operational contexts, defined as the source domain. However, these methods typically assume that fault categories between source and target domains are the same. This assumption is invalid in real applications where the target domain often contains fewer fault types than the source. Such scenarios may introduce negative transfer caused by irrelevant source domain categories. To solve this problem, a novel dual-weight partial domain adaptation (DW-PDA) method is proposed, which implements selective knowledge transfer by the class-wise adversarial learning for category-aware feature alignment, and a dual-weight mechanism for dynamic sample selection. Experimental results on both HVAC air-side and water-side datasets demonstrate the effectiveness of the DW-PDA in partial domain adaptation scenarios. Compared to the global domain adaptation methods, the average accuracy improvement of the DW-PDA is 30.87% and 25.26% in AHU and chiller tasks, respectively.

Abstract

Compressed air energy storage is considered one of the most promising large scale energy storage technologies due to its advantages of large scale, high efficiency and long service life. Isobaric system, compared to isochoric system, can effectively improve the energy density of the system. Therefore, this paper proposes an isobaric system with water compensation. The thermodynamic model of the system is established, and it is used to analyze the effects of molten salt low temperature, compressor polytropic efficiency, turbine polytropic efficiency, gas storage pressure and depth on the system performance. The analysis results show that within a certain range, increasing the molten salt low temperature, compressor polytropic efficiency, and turbine polytropic efficiency can effectively improve the round-trip efficiency of the system. Moreover, the deeper the underground depth, the higher the round-trip efficiency and energy density of the system. The increase in gas storage pressure can enable the system to achieve a higher energy density.

Abstract

Shared Socioeconomic Pathways (SSPs), which provide national-level scenarios for population and gross domestic product (GDP), are useful for analyzing future energy demand. However, their spatial resolution is too coarse to apply them for urban energy management, which typically requires data at the building or district level. Unfortunately, it is unclear how to downscale GDPs whose spatially detailed historical data are unavailable. To address this issue, this study develops a method for downscaling GDP scenarios from the national level to the building level while ensuring consistency with SSPs. We leverage micro-scale auxiliary variables and apply the power law, which captures the rank-size relationships underlying gross products, to accurately project the GDPs for each SSP. The proposed method is applied to project the building-level GDP under the SSPs for Yokohama, Japan. Furthermore, we assess the model’s applicability to other regions.

Abstract

In high-density cities, anthropogenic heat (AH) emissions are an important factor affecting the urban heat island effect. Especially in summer, the heat stress on the outdoor thermal environment caused by AH emissions cannot be ignored. Due to the oversimplification of AH emissions in previous studies, the spatial differences in the role of different types of AH emissions in the urban thermal environment are not well understood. Therefore, this study proposes an interpretable machine learning framework to model and predict AH emissions and their thermal impacts in dense urban areas, with a focus on quantifying the differential responses of land surface temperature (LST) to sector-specific AH emissions (e.g., buildings, industry, transportation). The results demonstrate that the ExtraTreesRegressor model achieves optimal performance, with a coefficient of determination (R²) of 0.8. The Shap analysis reveals that, among the seven types of selected urban factors, NDVI, Building-related AH, and Industry-related AH are the key driving factors of LST across Shenzhen, China. Sensitivity experiments demonstrate that any reduction in the AHE intensity results in a corresponding decrease in the spatially averaged LST. Under an 80% reduction in AHE, the mean LST in Shenzhen decreased by 0.57 °C (buildings), 0.46 °C (industry), and 0.23 °C (transportation). Furthermore, the cooling response to AH reduction varied spatially across Shenzhen. A reduction in building or industry-related AH emissions induced significant cooling areas in inland districts, including Guangming, Longhua, Pingshan, and parts of Futian. In contrast, coastal districts such as Nanshan and Futian exhibited more pronounced cooling effects when transportation-related AH emissions were reduced. This study elucidates the mechanistic response of urban thermal environments to spatially heterogeneous and source-specific AH emissions,providing novel insights and a theoretical foundation for developing targeted urban heat mitigation strategies.

Abstract

Compressed carbon dioxide energy storage has received widespread attention as an effective solution to address the intermittency of renewable energy sources. This article proposes a novel compressed carbon dioxide energy storage system with a flexible gas warehouse compartment, which stores carbon dioxide at ambient temperature and pressure. A mathematical model is established and a parameter analysis is conducted to evaluate the thermodynamic and economic performance of the proposed system. The results indicate that increasing the mechanical efficiency of the compressor and expander can achieve lower electricity costs and higher round-trip efficiency. By reducing the pinch temperature of the heat exchanger, higher round-trip efficiency and lower electricity costs can be achieved.

Abstract

Integrated Energy Systems (IES) play a crucial role in promoting multi-energy complementarity and enhancing overall energy utilization efficiency. However, the intermittency of renewable energy sources and the stochastic nature of load demand pose significant challenges to system dispatch. To address these issues, this study proposes an economic dispatch method for renewable-integrated IES based on Deep Reinforcement Learning (DRL). The system components, including internal combustion engines, absorption chillers, and battery energy storage systems, are modeled, and the dispatch problem is formulated as a Markov Decision Process (MDP). Two DRL algorithms, Deep Q-Network (DQN) and Twin Delayed Deep Deterministic Policy Gradient (TD3), are employed to train optimal control strategies. Experimental results demonstrate that both algorithms achieve efficient system dispatch, with TD3 exhibiting superior convergence speed and overall performance. The proposed approach does not rely on explicit uncertainty modeling and can adaptively respond to fluctuations in renewable generation and load demand, thereby improving the economic efficiency and robustness of system operation.