The operation and maintenance (O&M) of buildings plays an important role in ensuring that the buildings work normally, as well as reducing the damage caused by functional errors. There are obvious problems in the traditional O&M modality, and an effective way to solve them is to make the model smarter. In this paper, a digital twin framework for building operation is proposed, which consists of two key components: a digital twin O&M model and a machine learning algorithm. The process of establishing the digital twin model is introduced in detail, and the method is explained according to the structure, equipment, and energy consumption characteristics of the model. A mechanism of fusing the digital twin and machine learning algorithm is proposed and the prediction process based on an artificial neural network (ANN) is shown. Finally, based on a systematic summary of the modeling process and fusion mechanism, the development path and overall structure of the intelligent O&M system utilizing digital twins is proposed.

In traditional construction safety assessment, it is difficult to describe the safety status of different construction stages. To solve this problem, this paper proposes a digital twin modeling theory for construction safety assessment. Firstly, this paper analyzes the requirements of a digital twin model. Secondly, the required information is collected by IoT. Finally, the DT model is established based on the collected information. This DT model analyzes the collected information by ML, which aims to conducting the assessments of construction safety. To verify this method, this paper analyzes the vault settlement during tunnel construction. The analysis results show that the DT model can predict the settlement value with high accuracy. Moreover, the safety state is assessed dynamically based on the settlement value by DT.

The most negative effects caused by earthquakes are the damage and collapse of buildings. Seismic building retrofitting and repair can effectively reduce the negative impact on post-earthquake buildings. The priority to repair the construction after being damaged by an earthquake is to perform an assessment of seismic buildings. The traditional damage assessment method is mainly based on visual inspection, which is highly subjective and has low efficiency. To improve the intelligence of damage assessments for post-earthquake buildings, this paper proposed an assessment method using CV (Computer Vision) and AR (Augmented Reality). Firstly, this paper proposed a fusion mechanism for the CV and AR of the assessment method. Secondly, the CNN (Convolutional Neural Network) algorithm and gray value theory are used to determine the damage information of post-earthquake buildings. Then, the damage assessment can be visually displayed according to the damage information. Finally, this paper used a damage assessment case of seismic-reinforced concrete frame beams to verify the feasibility and effectiveness of the proposed assessment method.