Since the 18th National Congress of the Communist Party of China, a series of policies have been studied and issued around the development of digital agriculture, which provides unprecedented policy support and institutional guarantee for the development of digital agriculture. However, due to the late start of digital technology in China, compared with some mature developed countries, the gap is relatively large. At present, there are few domestic enterprises doing digital well control terminal and Internet of things big data platform technology, and the technological development of enterprises is relatively imperfect and imperfect. Therefore, the digital well control terminal and Internet of things big data platform technology on the market have great defects in performance and quality, it is difficult to achieve high-efficiency mass production, basically still in the trial stage, so far it is difficult to promote the use. At present, the product design of foreign products can not meet the requirements of farming in the vast farming areas in China, and the import price of foreign products is generally higher. Therefore, the agricultural condition monitoring system based on multi-modal data fusion came into being. This work uses water level sensor, water flow sensor, soil moisture sensor, light sensor, temperature and humidity sensor and wind sensor to capture relevant sensing information, and then transmits the data information to STM32. An improved multi-sensor data fusion algorithm based on the combination of D-S evidence theory and BP neural network is built in STM32, after the data collected by different types of sensors are fused by BP data, the characteristic values of farmland environment are obtained. On this basis, the basic probability distribution function and evidence fusion rules of D-S evidence theory are designed. The analyzed and processed data is transmitted wirelessly through the BC95 module, the data is uploaded to the Tengxun cloud server for data storage, and finally the data is displayed through the WEB side and the APP side. This system realizes the networking of farmland motor wells, digitizes the status of motor well supporting equipment, realizes all-weather, all-round, and full-node visual monitoring, improves the life of motor wells, guarantees the results of government investment, promotes the modernization of agriculture and rural areas, and enhances the economic and social development of agriculture. The ability to support and guarantee has continuously improved the living standards of farmers.