Building energy assessment models are considered to be one of the most informative methods in building energy efficiency design, and most of the current building energy assessment models have been developed based on machine learning algorithms. Deep learning models have proved their effectiveness in fields such as image and fault detection. This paper proposes a deep learning energy assessment framework with interpretability to support building energy efficiency design. The proposed framework is validated using the Commercial Building Energy Consumption Survey dataset, and the results show that the wrapper feature selection method (Sequential Forward Generation) significantly improves the performance of deep learning and machine learning models compared with the filtered (Mutual Information) and embedded (Least Absolute Shrinkage and Selection Operator) feature selection algorithms. Moreover, the Deep Forest model has an R² of 0.90 and outperforms the Deep Multilayer Perceptron, the Convolutional Neural Network, the Backpropagation Neural Network, and the Radial Basis Function Network in terms of prediction performance. In addition, the model interpretability results reveal how the features affect the prediction results and the contribution of the features to the energy consumption in a single building sample. This study helps building energy designers assess the energy consumption of new buildings and develop improvement measures.

The chemical industry is a typical high-carbon emitting industry, and achieving the goal of net zero emissions by 2050 is challenging. Therefore, metal chemical enterprises have to explore a special path of low-carbon development. This article conducted a case study on a Chinese metal chemical production enterprise with a processing scale of 28,000 t/year. Starting from the analysis of energy consumption carbon emissions, this article used available statistical data at the enterprise level to build a carbon emission estimation model for the enterprise combining different emission categories. Moreover, we also calculated the carbon emissions and carbon emission intensity of the enterprise from 2014 to 2022. Further quantitative analyses on the impact of production scale, energy efficiency, energy structure, and emission coefficient on carbon increment were also conducted using a logarithmic mean divisia index (LMDI) model. The results showed that the reduction in carbon emissions of the enterprise during the research period was due to the improvement of energy efficiency, while the production scale and energy structure served as important driving factors. Based on the results, this article proposes some policy suggestions on the future direction and focus of the enterprise’s carbon reduction work.