Abstract: The efficient aggregation and shared application of real-time hydrological and rainfall data from reservoir clusters serve as a critical data foundation for scientifically conducting basin-wide flood forecasting and warning, dynamically simulating flood routing analyses, and precisely formulating flood control operation plans. In response to new situations, tasks, and requirements in flood and drought management, this study investigates the information management needs of reservoir clusters in the river basin and analyzes the current state of information aggregation for reservoir clusters in the upper Changjiang River. Addressing characteristics such as data heterogeneity, diverse sources, and variable quality, a design and development scheme for a data management platform for Changjiang River Basin reservoir clusters is proposed, which incorporates big data analytics technologies. The platform adopts the SpringBoot+Vue technology stack and a microservices architecture, designing a layered framework comprising data, business, and user layers. It integrates multi-source heterogeneous data and enables standardized interaction, establishing a dual-dimensional anomaly detection mechanism combining "threshold-based judgment and machine learning." By employing a unified indicator system and a "site-specific strategy" approach, the platform utilizes the XGBoost algorithm to achieve accurate identification of anomalies in hydrological data. Core functional modules, including integrated monitoring, anomaly detection, information maintenance, and file management, have been developed to ensure real-time data management and sharing. The platform has successfully realized the integration and real-time management of multi-source heterogeneous data from Changjiang River Basin reservoir clusters. The XGBoost model demonstrates an anomaly detection precision of 0.79, a recall rate of 0.8, and a false positive rate of only 0.002 1, meeting practical operational requirements. Through its standardized architecture and intelligent detection technology, the platform effectively enhances data quality and sharing efficiency. It provides a foundational platform for further improving cross-departmental information aggregation and management capabilities within the basin and supports flood and drought disaster prevention and mitigation efforts in the Changjiang River Basin.