With the development of new power systems, transmission monitoring text data presents the characteristics of large volume and fast growth rate, and because of the privatization of industry data transmission protocols, the performance of data retrieval is low. This affects the real-time decision analysis of transmission lines. An intelligent retrieval model of power transmission monitoring data based on MapReduce is proposed. First, this paper innovates and improves the SimHash algorithm to achieve efficient extraction of retrieval vectors for transmission line online monitoring text data, and introduces multi-attribute decision-making and comprehensive scoring mechanisms to achieve precise retrieval of target data and improve data retrieval accuracy and recall. Second, from the characteristics of large data volume and fast growth rate, a power data retrieval model based on MapReduce is designed. Finally, the retrieval accuracy, recall rate and retrieval efficiency of the proposed method are verified through comparative analysis of power grid examples.