The geographical locations of nuclear power plants are scattered, and massive amounts of time series data is generated during production and operation. The safety, reliability, and efficient management of time series data poses significant challenges to the power plants. In order to solve the problem of data islands and improve the management level and sharing efficiency of group level time series data, this paper designs the Time Series Lake Platform: TSLP for nuclear power industrial Internet scenario, which can manage multiple instances of time series databases and the collection, storage, and access processes of time series data in nuclear power plants and data centers, meeting the requirements of security and reliability. Designed a cloud-edge integrated time series data processing architecture, laying the foundation for data aggregation and governance; Conducting research on parallel data transmission and consumption through multiple channels to avoid the occurrence of data loss or severe delay caused by a single point of transmission link and consumption program; By using functions such as connection management, configuration management, data management, operation and maintenance management, data analysis, the low-frequency and high-frequency time series data from various nuclear power plants different and units are classified and managed according to business attributes, improving management efficiency and enhancing data retrieval efficiency; Research unified nuclear power time series data services, including authentication and concurrent access control methods to time series data, ensuring the security and stability of data access. The TSLP proposed in this paper can serve as the underlying data platform for digital nuclear power, providing professional support for the collection, storage, access and analysis of time series data for digital applications.