With the development of the times, how to share data safely on the basis of protecting users' privacy has attracted extensive attention. As a new machine learning technology, federated learning can use the local data set of nodes for distributed model training, and only share the model without uploading the original training data during the training update process, so as to achieve the safe sharing of the model. However, the data provider will not share the local data out of concern about the risk of data security and privacy disclosure, resulting in data sharing failure; There may be malicious workers in federal learning that will destroy learning, affect the overall model, and lead to unreliable data sharing; In the process of model updating, intermediate parameters may be stolen by attackers, leading to the privacy disclosure of data providers.To address the above issues, this paper introduces blockchain technology to further study data sharing and privacy protection in federated learning.

With the rapid development of new technologies such as the Internet of Things, artificial intelligence and cloud computing, human beings have entered the era of big data, and with it comes the explosive growth of data. By using new technologies to analyze and process data, the hidden values in the data can be unearthed. Therefore, it is essential to establish a fair and secure data trading bridge between the generation and use of data. Currently, most of the data trading platforms are based on centralized institutions, but centralized trading platforms have problems such as single point of failure, privacy disclosure, transaction opacity, data resale and so on.Blockchain, as an emerging technology in recent years, provides new ideas for data trading due to its decentralized, traceable and tamper-proof technical characteristics. Relevant scholars propose blockchain-based data trading scheme, and although it solves some problems in centralized data trading, the existing scheme still suffers from unguaranteed data quality, privacy leakage, unreasonable data pricing, etc.A data trading model based on quality detection and negotiated pricing is proposed to solve the problems of single point of failure, leakage of user privacy, unguaranteed data quality, and unreasonable data pricing in existing data trading schemes. The model combines smart contracts, cryptography, trusted hardware and other technologies to accomplish fair data transactions while protecting user privacy. By building a test environment to verify the feasibility and performance analysis of the model, the experimental results show that the model not only solves the above problems, but also ensures the fairness and security of the whole transaction process.

In this paper, the test data and analysis results are used to guide the correction of late blasting parameters, adjust the scope of static excavation area and determine the blasting vibration velocity standard. The rock mass in the static excavation area becomes loose due to vibration after the excavation face blasting operation of the rear tunnel. The hydraulic crushing hammer is used to chisel out the surrounding rock in the static excavation area faster, which has little influence on the construction progress. At the same time, the static excavation method with excavator and crushing hammer has high controllability, which ensures that the excavation of the rear tunnel does not invade the initial support contour of the first tunnel, and reduces the construction disturbance to the initial support of the adjacent position of the first tunnel. Avoid the construction of the rear tunnel leading to the first hole lining structure cracking and instability, reduce the amount of the first hole lining cracks, improve the safety of the construction process, to ensure the project duration and quality requirements.

With the rapid development of the construction industry in China, people have higher and higher requirements for the sustainable development of the environment. As a new building form, prefabricated buildings have the characteristics of standardization, industrialization and efficiency, and have been widely used in the process of engineering construction. The application of BIM technology in the construction management of prefabricated building projects can realize the information integration management of the whole process of prefabricated building construction, improve the cooperative work efficiency of all participants, and give full play to the advantages of prefabricated buildings. This paper expounds the application of BIM technology in the construction management of prefabricated buildings, and discusses the importance of BIM technology.